Methodological issues in spatial microsimulation modelling for small area estimation

Small area estimation is the method of estimating reliable statistics at the small geographical area or a spatial micropopulation unit. Reliable statistics of interest at small area levels cannot be ordinarily and directly produced, due to certain limitations of the available data. For instance, a suitable sample that contains enough representative observations is typically not available for all small areas from national level survey data. A basic problem with national or state level surveys is that they are not designed for efficient estimation for small areas (Heady et al., 2003). In practice, small area estimates from these national sample surveys are statistically unreliable, due to sample observations being insufficient, or in many cases non-existent, where the domain of interest may fall outside the sample domains (Tanton, 2007). Given typical time and money constraints, it is usually impossible to conduct a sufficiently comprehensive survey to get enough data from every small area we are interested in.

Nowadays indirect modelling approaches of small area estimation, such as spatial microsimulation models (SMMs), have received much attention, due to their usefulness and the increasing demand for reliable small area statistics from both private and public level organisations. In these approaches, one uses data from similar domains to estimate the statistics in a particular small area of interest, and this ‘borrowing of strength’ is justified by assuming a model that relates the small area statistics (Meeden, 2003). Typically, indirect small area estimation is the process of using statistical models and/or geographic models to link survey outcome or response variables to a set of predictor variables known for small areas, in order to predict small area estimates. As a result of inadequate sample observations in small geographic areas, the conventional area-specific direct estimates may not provide enough statistical precision. In such a situation, an indirect model-based method can produce better results.

Most of the review articles in small area estimation have highlighted the methodologies, which are fully based on various statistical models and theories (for example, Ghosh and Rao, 1994; Rao, 1999; Pfeffermann, 2002; Rao, 2002; Rao, 2003a). However another type of technique called ‘spatial microsimulation modelling’ has been used in providing small area estimates during the last decade (for instance, Williamson et al., 1998; Ballas et al., 2003; Taylor et al., 2004; Brown and Harding, 2005; Chin et al., 2005; Ballas et al., 2006; Chin and Harding, 2006; Cullinan et al., 2006; Lymer et al., 2006; Anderson, 2007; Chin and Harding, 2007; King, 2007; Tanton, 2007). The SMMs are based on geographic and economic theories, and their methodologies are quite different from other statistical approaches. Although these approaches are frequently used in social and economic analysis, and seem to be a robust and rational indirect modelling tool, the mechanisms behind them are not always well documented. Also there are some important methodological issues where more research should improve the performance of SMMs and help in the validation of their estimates.

This paper provides a brief synopsis of the overall methodologies for small area estimation and explicitly addresses some vital methodological issues of spatial microsimulation modelling, with a particular emphasis given to the reweighting techniques. It also proposes a new approach in the SMM methodologies. An application of the generalised regression based reweighting technique discussed in this article is studied by Tanton and Vidyattama under distinct features of the applicable data. This contribution is part of this special issue as well.

There are 5 sections within this paper. In Section 2, a diagramic representation of the overall methodologies in small area estimation is provided with a synopsis of various direct and indirect statistical model based estimations, and highlights of spatial microsimulation modelling. In Section 3, some vital methodological issues in spatial microsimulation modelling are addressed, which include theories and numerical solutions of different reweighting techniques. In Section 4, a comparison between two reweighting techniques is presented with a new methodology for generating small area microdata. Finally, conclusions are given in Section 5.

A diagramic representation of the overall methodologies for small area estimation is depicted in Figure 1. Traditionally there are two types of small area estimation – direct and indirect estimation. Direct small area estimation is based on survey design and includes three estimators called the Horvitz-Thompson estimator, GREG estimator and modified direct estimator. On the other hand, indirect approaches of small area estimation can be divided into two classes – statistical and geographic approaches. The statistical approach is mainly based on different statistical models and techniques. However, the geographic approach uses techniques such as SMMs.

Figure 1

Download asset Open asset

It is noted that implicit model based statistical approaches include three types of estimators, which are synthetic, composite and demographic estimators. Whereas, there are also three kinds of explicit models categorized as area level, unit level and general linear mixed models. Based on the type of study researchers are interested in, each of these models is widely applied to obtained small area indirect estimates by utilising the (empirical-) best linear unbiased prediction (E-BLUP), empirical Bayes (EB) and hierarchical Bayes (HB) methods. A very brief synopsis of different direct and indirect statistical model based small area estimation techniques is presented in Table 1 (for details, see Rahman, 2008a).

However, in contrast to statistical approaches, the geographic approach is based on microsimulation models, which are essentially creating synthetic/simulated micro-population data to produce ‘simulated estimates’ at small area level. To obtain reliable microdata at small area level is the key task for spatial microsimulation modelling. Synthetic reconstruction and reweighting are two commonly used methods in microsimulation, and each of them is stimulated by different techniques to produce simulated estimators. As the main objective of this paper is to discuss the methodological issues of spatial microsimulation modelling, the subsequent sections will encompass those methods for explicit treatments.

As mentioned calculating statistically reliable population estimates in a local area using survey microdata is challenging, due to the lack of enough sample observations. To create a synthetic spatial microdata set is one of the possible solutions. Simulation based methods can deal with such a problem by (re)weighting each respondent in the survey data, to create the synthetic spatial microdata. However, it is not easy process to create reliable spatial microdata. Complex methodologies are associated with the process. This section presents some of the vital methodological issues in spatial microsimulation modelling.

3.2 GREGWT theory: How does it generate new weights?

The GREGWT approach of reweighting is an iterative generalised regression algorithm written in SAS macros. Let us assume that a finite population is denoted by Ω = {1,2,…, k,…, N}, and a sample s (s ⊆ Ω) is drawn from Ω with a given probability sampling design p(.). Suppose the inclusion probability Π_k = Pr(k ∈ s) is a strictly positive and known quantity. Now for the elements k ∈ s, let (y_k,x_k) be a set of sample observations; where y_k is the value of the variable of interest for the k^th population unit and x^′_k = (x_k,1…k,x_k,j,…,x_k,p) is a vector of auxiliary information associated y’_k. Note that data for a range of auxiliary variables should be available for each unit of a sample s. In a particular case, suppose for an auxiliary variable j, the element x_k,j = 1 in x_k if the k^th individual is not in workforce, and x_k,j = 0 ‘otherwise’. Thus the number of individuals in the sample who are not in the workforce is given by

(1) ${\displaystyle \sum }_{k\in s}{x}_{k,j}.$

If the given sampling design weights are d_k = 1/Π_k (k ∈ s) then the sample based population totals of auxiliary information,

(2) ${\widehat{t}}_{x,s}={\displaystyle \sum _{k\in s}{d}_k{x}_k}$

can be obtained for a p-elements auxiliary vector x_k. But the true value of the population total of the auxiliary information T_x should be known from some other sources such as from the census or administrative records. In practice, ${\widehat{t}}_{x,s}$ is far from T_x when the sample s is a bad or poorly representative of the population.

For obtaining a more reliable small area estimate of population total of the variable of interest, we have to generate a new set of weights w_k for k ∈ s, for which the calibration equation

(3) ${\displaystyle \sum _{k\in s}{w}_k{x}_k=T_x}$

must be fitted and the new weights w_k will be as close as possible to d_k.

The distance measure used in the GREGWT algorithm is known as truncated Chi-squared distance function and it can be defined as

(4) ${\stackrel{{\chi }^2}{G}}_k=\frac{{(w_k-d_k)}^2}{2d_k};\text{for\hspace{0.17em}}{L}_k\le \frac{w_k}{d_k}\le U_k$

where L_k and U_k are pre specified lower and upper bounds respectively for each unit k ∈ s.

For a simple special case the total of this type of distance measure can be defined as

(5) $D=\frac{1}{2}{\displaystyle \sum _{k\in s}\frac{{\left(w_k-d_k\right)}^2}{d_k}}.$

Hence the Lagrangean for the Chi-squared distance function is

(6) $L=\frac{1}{2}\underset{k\in s}{{{\displaystyle \sum }}^{\text{}}}\frac{{(w_k-d_k)}^2}{d_k}+\underset{j=1}{\overset{p}{{{\displaystyle \sum }}^{\text{}}}}(T_{x,j}-\underset{k\in s}{{{\displaystyle \sum }}^{\text{}}}{w}_k{x}_{k,j})$

where λ_j (j=1,2,…,p) are the Lagrange multipliers, and T_x,j is the j^th element of the vector of true values of known population total for the auxiliary information, T_x.

By differentiating (6) with respect to w_k and then applying the first order condition, we have

(7) $\frac{\partial L}{\partial w_k}=\left(\frac{w_k-d_k}{d_k}\right)-{\displaystyle \sum }_{j=1}^p{\lambda }_j{x}_{k,j}=0$

for k ∈ s ⊆ Ω, along with the p^th (j=1,2,…p) constraints conditions in equation (3). As earlier, for a simple representation it is convenient to write

(8) $x_k^{\prime }\lambda ={{{\displaystyle \sum }}^{\text{}}}^{\text{}}{\lambda }_{{}_j}{x}_{k,j}.$

Hence the new weights can be formulated as

(9) $w_k=d_k+d_k{x}_k^{\prime }\lambda .$

To find the values of the Lagrange multipliers, the equation (9) can be rearranged in a convenient form. After multiplying the equation by x_k and then summing over k it can be written as

(10) ${\displaystyle \sum }_{k\in s}{w}_k{x}_k={\displaystyle \sum }_{k\in s}{d}_k{x}_k+{\displaystyle \sum }_{k\in s}{d}_k{x}_k{x}_k^{\prime }\lambda .$

Now since ${\displaystyle \sum _{k\in s}{d}_k{x}_k}={\widehat{t}}_{x,s}$ and ${\displaystyle \sum _{k\in s}{w}_k{x}_k=T_x}$ are known, the above equation can be expressed as

(11) $\left({\displaystyle \sum }_{k\in s}{d}_k{x}_k{x}_k^{\prime }\right)\lambda =T_x-{\widehat{t}}_{x,s}$

where the summing term in brackets is a p × p symmetric-square matrix. If the inverse of this matrix exists, the vector of Lagrange multipliers can be obtained by the following equation

(12) $\lambda ={\left(\underset{k\in s}{{\sum }^{\text{}}}{d}_k{x}_k{x}_k^{\prime }\right)}^{-1}(T_x-{\widehat{t}}_{x,s});\text{\hspace{0.17em}for\hspace{0.17em}}\left|\underset{k\in s}{{{\displaystyle \sum }}^{\text{}}}{d}_k{x}_k{x}_k^{\prime }\right|\ne 0.$

Hence using the resulting values of Lagrange multipliers, λ, one can easily calculate the new weights w_k from equation in (9). Moreover to minimize the truncated Chi-squared distance function in (4), an iterative procedure known as the Newton-Raphson method (appendix A) is used in the GREGWT program (Bell, 2000). It adjusts the new weights in such a way that minimises equation (4) and produces generalised regression estimates or synthetic estimates of the variable of interest.

Explicit numerical solution for a hypothetical data An explicit numerical solution of the above very simple case theory is given here. Let x_k,j is the j^th auxiliary variable linked with k^th sample unit for which true population values T_x are available from census or other administrative records. Suppose in a hypothetical dataset, observations of 25 sample units for a set of 5 auxiliary variables such as age (1 = 16–30 years and 0 = ‘otherwise’), sex (1 = female and 0= male), employment (1 = unemployed and 0 = ‘otherwise’), income from unemployment benefits (in real unit values 0, 1, 2, 3, 4 and 5) and location (1 = rural and 0 = urban) are available, and its associated auxiliary information matrix, sample design weights and the known population values vector are accordingly given as -

(13) $X=[x_{k,j}^{\prime }]=\left[\begin{array}{ccccc}1& 1& 0& 0& 0\\ 1& 0& 1& 3& 1\\ 0& 0& 1& 2& 1\\ 1& 1& 1& 5& 0\\ 0& 1& 0& 0& 1\\ 0& 0& 1& 1& 0\\ 0& 0& 0& 0& 1\\ 1& 0& 1& 4& 0\\ 0& 1& 0& 0& 1\\ 1& 0& 0& 0& 1\\ 0& 1& 1& 1& 0\\ 1& 1& 1& 3& 1\\ 1& 0& 1& 2& 1\\ 0& 0& 1& 5& 1\\ 0& 1& 1& 4& 0\\ 0& 0& 0& 0& 0\\ 1& 0& 1& 3& 1\\ 0& 1& 0& 0& 0\\ 0& 0& 1& 2& 1\\ 1& 0& 1& 4& 0\\ 0& 0& 0& 0& 1\\ 0& 0& 1& 5& 1\\ 1& 1& 0& 0& 0\\ 0& 1& 1& 1& 0\\ 1& 0& 0& 0& 1\end{array}\right],d=(d_k)\begin{array}{c}\left[\begin{array}{c}4\\ 5\\ 6\\ 5\\ 3\\ 4\\ 6\\ 4\\ 5\\ 3\\ 5\\ 4\\ 3\\ 6\\ 4\\ 5\\ 6\\ 3\\ 6\\ 4\\ 5\\ 3\\ 5\\ 4\\ 3\end{array}\right]\end{array}\text{\hspace{0.17em}and\hspace{0.17em}}{T}_x=\left(\begin{array}{c}50\\ 45\\ 70\\ 200\\ 65\end{array}\right)$

Note: the 1^st row of matrix X represents a sample unit of age between 16 to 30 years, female, in ‘otherwise’ employment categories that is may be in labour force or employed, with a real unit value of income from unemployment is 0 dollar, and living in an urban area.

Now we have to estimate ${\widehat{t}}_{x,s}$ and the inverse matrix of $\left({\displaystyle \sum }_{k\in s}{d}_k{x}_k{x}_k^{\prime }\right)=\text{A}(say)$.

By using mathematical formulas one can easily obtain,

(14) ${\widehat{t}}_{x,s}={\left({\displaystyle \sum }_{k=1}^{25}{d}_k{x}_{k,1},{\displaystyle \sum }_{k=1}^{25}{d}_k{x}_{k,2},{\displaystyle \sum }_{k=1}^{25}{d}_k{x}_{k,3},{\displaystyle \sum }_{k=1}^{25}{d}_k{x}_{k,4},{\displaystyle \sum }_{k=1}^{25}{d}_k{x}_{k,5}\right)}^{\prime }={\left(\begin{array}{ccccc}46& 42& 69& 206& 64\end{array}\right)}^{,},\text{\hspace{0.17em}and}$

(15) $A=\left[\begin{array}{ccccc}A11& A12& A13& A14& A15\\ A21& A22& A23& A24& A25\\ A31& A32& A33& A34& A35\\ A41& A42& A43& A44& A45\\ A51& A52& A53& A54& A55\end{array}\right]=\left[\begin{array}{ccccc}46& 18& 31& 108& 24\\ 18& 42& 22& 62& 12\\ 31& 22& 69& 206& 39\\ 108& 62& 206& 750& 120\\ 24& 12& 39& 120& 64\end{array}\right]$

where Ajj = ${\displaystyle \sum _{k=1}^{25}{d}_k,x_{k,j}{x}_{k,j}^{\prime }={\displaystyle \sum _{k=1}^{25}{d}_k{x}_{k,j}^2}}$ and Aij = ${\displaystyle \sum _{k=1}^{25}{d}_k,x_{k,i}{x}_{k,j}^{\prime }}$; for all i,j (=1,2,3,4,5) and i ≠ j.

The inverse matrix of A = $\left({\displaystyle \sum _{k\in s}{d}_k{x}_k{x}_k^{\prime }}\right)$ can be obtained as ${\text{A}}^{-1}={\left({\displaystyle \sum _{k\in s}{d}_k{x}_k{x}_k^{\prime }}\right)}^{-1}$

(16) $=\left[\begin{array}{ccccc}0.03661582& -0.00901288& 0.00228602& -0.00429437& -0.00538212\\ -0.00901288& 0.03088625& -0.01214961& 0.00183273& 0.00155596\\ 0.00228602& -0.01214961& 0.09100053& -0.02239201& -0.01204764\\ -0.00429437& 0.00183273& -0.02239201& 0.00794951& 0.00000656\\ -0.00538212& 0.00155596& -0.01204764& 0.00000656& 0.02468079\end{array}\right]$

Then by using the results in relationship (12), the Lagrange multipliers should be calculated for this simple particular example as: λ′ = (0.14209475, 0.03501717, 0.18600019, −0.08176176, −0.00426682) .

Now using this result in equation (9), the new weights or calibrated weights for the Chi-squared distance measure can be easily obtained. The calculated new weights and its distance measures to the sample design weights are given in Table 3. For the 16th unit of our hypothetical data, the new weight remains unchanged to the sampling design weight due to the fact that all entries for this unit are zero. However this is very rare in GREGWT reweighting.

Table 3

dk	wk	wk −dk	$\stackrel{{\chi }^2}{G_k}$
4	4.70844769	0.70844769	0.06273727
5	5.39271424	0.39271424	0.01542245
6	6.10925911	0.10925911	0.00099480
5	4.77151662	−0.22848338	0.00522047
3	3.09225105	0.09225105	0.00141838
4	4.41695372	0.41695372	0.02173130
6	5.97439907	−0.02560093	0.00005462
4	4.00419164	0.00419164	0.00000220
5	5.15375174	0.15375174	0.00236396
3	3.41348379	0.41348379	0.02849481
5	5.69627800	0.69627800	0.04848031
4	4.45424007	0.45424007	0.02579175
3	3.48091381	0.48091381	0.03854635
6	4.63754748	−1.36245252	0.15468974
4	3.57588131	−0.42411869	0.02248458
5	5.00000000	0	0
6	6.47125708	0.47125708	0.01850694
3	3.10505151	0.10505151	0.00183930
6	6.10925911	0.10925911	0.00099480
4	4.00419164	0.00419164	0.00000219
5	4.97866589	−0.02133411	0.00004551
3	2.31877374	−0.68122626	0.07734487
5	5.88555961	0.88555961	0.07842158
4	4.55702240	0.55702240	0.03878424
3	3.41348379	0.41348379	0.02849481
		TAD = 9.21152591	D = 0.67286721

In addition, the total absolute distance (TAD) indicates higher quantity. While absolute distance has a higher value, the corresponding Chi-squared distance measure also indicates a higher value. However the fluctuations within absolute distances are remarkable compared to Chi-squared distance measures (see in Figure 2).

Figure 2

Download asset Open asset

Furthermore, when the TAD will zero the total Chi-squared distance will also be zero, and in that situation the calibrated weights will remain same as the sampling design weights which indicates the sample data are fully representative to the small area population. Moreover, it is interesting to note that the values of a set of new weights vary greatly with the changing values of vector for differences between ${\widehat{t}}_{x,s}$ and T_x. These differences can come from the poorly representative data and/or the chosen benchmarks. Four random alternative cases of difference vectors

(17) $\begin{array}{c}C1={\left[4,3,1-6,1\right]}^{\prime },\\ C2={\left[8,3,1-6,1\right]}^{\prime },\\ C3={\left[12,3,1-6,1\right]}^{\prime },\\ \text{and}C4={\left[4,3,1,2,1-6,1\right]}^{\prime },\\ \text{where}{C}_j=\left[T_x-{\widehat{t}}_{x,x}\right]\text{for}j=1,2,3,4,\end{array}$

have been considered and the resulting sets of new weights are plotted in Figure 3. The results show that the case C4 generates a more consistent set of new weights compared to the other cases. It is obvious that when the auxiliary information matrix provides quite rich sample data then the resulting difference vector between ${\widehat{t}}_{x,s}$ and T_x will be fairly close. Hence the resulting set of calibrated weights will produce more accurate estimates.

Figure 3

Download asset Open asset

3.3 Combinatorial optimisation reweighting approach

The combinatorial optimisation (CO) reweighting approach was first suggested in Williamson et al. (1998) as a new approach to create synthetic micro-populations for small domains. This reweighting method is mainly motivated towards selecting an appropriate combination of households from survey data to attain the known benchmark constraints at small area levels using an optimization tool. In the combinatorial optimisation algorithms, an iterative process begins with an initial set of households randomly selected from the survey data, to see the fit to the known benchmark constraints for each small domain. Then a random household from the initial set of combinations is replaced by a randomly chosen new household from the remaining survey data to assess whether there is an improvement of fit. The iterative process continues until an appropriate combination of households that best fits known small area benchmarks is achieved (Williamson et al., 1998; Voas & Williamson, 2000; Huang & Williamson, 2001; Tanton et al., 2007). The overall process involves five steps which are as follows:

1. collect a sample survey microdata file (such as CURFs in Australia) and small area benchmark constraints (for example, from census or administrative records);

2. select a set of households randomly from the survey sample which will act as an initial combination of households from a small area;

3. tabulate selected households and calculate total absolute difference from the known small area constraints;

4. choose one of the selected households randomly and replace it with a new household drawn at random from the survey sample, and then follow step 3 for the new set of households combination; and

5. repeat step 4 until no further reduction in total absolute difference is possible.

Note that when an array based survey data set contains a finite number of households it is possible to calculate all possible combinations of households. In theory, it may also be possible to find the set of households’ combination that best fits the known small area benchmarks. But, in practice, it is almost unachievable, due to computing constraints for a very large number of all possible solutions. For example, to select an appropriate combination of households for a small area with 150 households from a survey sample of 215789 households, the number of possible solutions greatly exceeds a billion (Williamson et al., 1998).

To overcome this difficulty, the combinatorial optimisation approach uses several ways of performing ‘intelligent searching’, effectively reducing the number of possible solutions. Williamson et al. (1998) provide a detailed discussion about three intelligent searching techniques: hill climbing, simulated annealing and genetic algorithms. Later on, to improve the accuracy and consistency of outputs, Voas and Williamson (2000) developed a ‘sequential fitting procedure’, which can satisfy a level of minimum acceptable fit for every table used to constrain the selection of households from the survey sample data. The following section will address the simulated annealing method only.

The simulated annealing method in CO

Simulated annealing, an intelligent searching technique for optimisation problems, has been successfully used in the CO reweighting process to create spatial microdata. The method is based on a physical process of annealing – in which a solid material is first melted in a heat bath by increasing the temperature to a maximum value at which point all particles of the solid have high energies and the freedom to randomly arrange themselves in the liquid phase. The process is then followed by a cooling phase, in which the temperature of the heat bath is slowly lowered. When the maximum temperature is sufficiently high and the cooling is carried out sufficiently slowly then all the particles of the material eventually arrange themselves in a state of high density and minimum energy. Simulated annealing has been used in various combinatorial optimisation problems (Kirkpatrick et al., 1983; van Laarhoven & Aarts, 1987; Williamson et al., 1998; Pham & Karaboga, 2000; Ballas, 2001).

The simulated annealing algorithm used in the CO reweighting approach was originally based on the Metropolis algorithm, which had been proposed by Metropolis et al. (1953). To simulate the evaluation to ‘thermal equilibrium’ of a solid for a fixed value of the temperature T the authors introduced an iterative method, which generates sequences of states of the solid in the following way. As mentioned in the book Simulated Annealing: Theory and Applications by van Laarhoven and Aarts (1987, p. 8):

“Given the current state of the solid, characterized by the position of its particles, a small, randomly generated, perturbation is applied by a small displacement of a randomly chosen particle. If the difference in energy, ∂E, between the current state and the slightly perturbed one is negative, that is, if the perturbation results in a lower energy for the solid, then the process is continued with the new state. If ∂E ≥ 0, then the probability of acceptance of the perturbed state is given by exp(∂E/K_BT). This acceptance rule for new states is referred to as the Metropolis Criterion. Following this criterion, the system eventually evolves into thermal equilibrium, that is, after a large number of perturbations, using the aforementioned acceptance criterion, the probability distribution of the states approaches the Boltzmann distribution, given as

(18) $p(\partial E)=\frac{1}{c(T)}\exp \left(-\frac{\partial E}{K_{B}T}\right)$

where c(T) is a normalizing factor depending on the temperature T and K_B is the Boltzmann constant.”

To search an appropriate combination of households from a survey dataset that best fits the benchmark constraints at small area levels is a combinatorial optimisation problem, and solutions in a combinatorial optimisation problem are equivalent to states of a physical annealing process. In the process of CO reweighting by simulated annealing algorithm, a combination of households takes the role of the states of a solid while the total absolute distance (TAD) function and the control parameter (for example, rate of reduction) take the roles of energy and temperature respectively. According to Williamson et al. (1998), change in energy becomes potential change in households’ combination performance (assessed by TAD) to meet the benchmarks, and temperature becomes a control for the maximum level of performance degradation (% of reduction) acceptable for the change of one element in a combination of households by a random element picks from the sample data. The control parameter is then lowered in steps, with the system being allowed to approach equilibrium for each step by generating a sequence of combinations by obeying the Metropolis criterion.

In addition, the algorithm is terminated for some small value of the control parameter, for which practically no deteriorations are accepted. Hence the normalizing constant which is depending on the controlling factor as well as Boltzmann constant can be dropped from the probability distribution. In this particular case we have the equation:

(19) $p(\partial E)=\exp \left(-\frac{\partial E}{T}\right).$

There are two important features of this probability equation described by Williamson et al. (1998). One is that the smaller the value of difference in energy, ∂E, the greater is the likelihood of a potential replacement being made in a combination. Another feature is that the smaller the value of controlling factor T, the smaller the change in performance likely to be accepted.

A typical simulated annealing algorithm is depicted in Figure 4. The overall process consists of a series of iterations in which random shifting is occurring from an existing solution to a new solution among all possible solutions. To accept a new solution as the base solution for further iteration, a test of goodness-of-fit based on TAD is consistently checked. The rules of the test are if the change of the difference in energy is negative, the newly simulated solution is accepted unconditionally; otherwise it is accepted satisfying the abovementioned Metropolis criterion. The simulated annealing algorithm may be able to avoid deceiving at local extremum in the solutions. Moreover, a solution or selected combination of households by this algorithm can generate real individuals living in actual households in a sense that individuals are from modelled outputs and not synthetically reconstructed (Ballas, 2001).

Figure 4

Download asset Open asset

An illustration of CO process for hypothetical data A simplified combinatorial optimisation process is depicted in Figure 5. It is noted that in this process when the total absolute difference (aforementioned TAD) in gregwt section) is equal to zero, the selection of households’ combination indicates the best fit. In other words, in this case the new weights give the actual households units from the survey sample microdata, which are the best representative combination. Thus it is a selection process of an appropriate combination of sample units, rather than calibrating the sampling design weights to a set of new weights.

Figure 5

Download asset Open asset

In this section, a comparison of the two reweighting methodologies is given with a new approach to the creation of synthetic spatial microdata.

4.2 Bayesian prediction approach of small area microdata simulation

A new system for creating synthetic spatial microdata is offered in this subsection. It is noted that after the sample survey, a finite population usually has two parts – which are observed units in the sample called data and unobserved sampling units in the population (Figure 6). Suppose Ω represents a finite population in which Ω _i (say) is the subpopulation of small area i. Now if s_i denotes the observed sample units in the i^th area then we have

(20) $s_i\cup {\overline{s}}_i={\Omega }_i\subseteq \Omega \text{\hspace{0.17em}for\hspace{0.17em}}{\forall }_i,$

Figure 6

Download asset Open asset

where ${\overline{s}}_i$ denotes the unobserved units in the small area population. Let y_ij represents a variable of interest for the j^th characteristic of the population at i^th small area. Thus we always have the estimate of population total at i^th small area

(21) $t_{y_i}={\displaystyle \sum _{j\in s_i}{y}_{ij}}+{\displaystyle \sum _{j\in {\overline{s}}_i}{y}_{ij}}.$

The main challenge in this process is to establish the link of observed data to the unobserved sampling units in the population. It is a kind of prediction problem, where a modeller tries to find a probability distribution of unobserved responses using the observed sample and the auxiliary data. The Bayesian methodology (see Ericson, 1969; Lo, 1986; Little, 2007; Rahman, 2008b) can deal with such a prediction problem.

The Bayesian prediction theory is very straightforward and mainly based on the Bayes’s posterior distribution of unknown parameters (Rahman 2008c). Let y be a set of observed units from a model with a joint probability density p(y|θ), in which θ is a set of model parameters. If a prior density of unknown parameters θ is g, the posterior density of θ for given y can be obtained by Bayes’s theorem and defined as p(y|θ) α p(y|θ) g(θ).

Now, if $\overline{y}$ is the set of unobserved units in a finite population, then under the Bayesian methodology its prediction distribution can be obtained by solving the integral

(22) $p\left(\overline{y}|y\right)\alpha {\displaystyle {\mathrm{\int }}_{\theta }p(\theta |y)p(\overline{y}|\theta )}d\theta ,$

where p($\overline{y}$|θ) is the probability density of unobserved units. Details of the Bayesian prediction theory for various regression models are given in Rahman (2008c).

For an i^th small area, let a multivariate linear model for the observed sample units

(23) $\begin{array}{l}{Y}_i=(y_1,y_2,\mathrm{...},y_{ni}{)}^{\prime }\\ \text{be}{Y}_i=X_i\beta +E_i\end{array}$

with errors distribution

(24) $E_i\sim T_{n_{i}p}\left(0,I_{n_i\times n_i},{\sum }_{p\times p},v\right),$

and for unobserved population units let

(25) $\begin{array}{l}{\overline{Y}}_i=({\overline{y}}_1,{\overline{y}}_2,\mathrm{\dots },{\overline{y}}_{N_i-n_i}{)}^{\prime }\\ be{\overline{Y}}_i={\overline{X}}_i\beta +{\overline{E}}_i\end{array}$

with errors distribution

(26) ${\overline{E}}_i\sim T_{(N_i-n_i)p}(0,I_{(N_i-n_i)\times (N_i-n_i)},{\displaystyle {\sum }_{p\times p}v}),$

where all the notations are as usual (see Rahman 2008c).

Applying the Bayesian prediction theory under a prior distribution

(27) $p(\beta ,\Sigma )\infty {[\Sigma ]}^{-(p+1)/2},$

we can derive the distribution of unobserved population units as

(28) $f({\overline{Y}}_i|Y_i,\beta ,\Sigma )=C{\left[S_{Y_i}+({\overline{Y}}_i-{\overline{X}}_i\widehat{\beta }{)}^{\prime }H({\overline{Y}}_i-{\overline{X}}_i\widehat{\beta })\right]}^{\frac{N_i-k}{2}}$

where

$\widehat{\beta }$ is the OLS of β,

(29) $S_{Y_i}=(Y_i-X_i\widehat{\beta }{)}^{\prime }(Y_i-X_i\widehat{\beta })$

(30) $H=\left[I-{\overline{X}}_i{\left(X_i^{\prime }{X}_i+{\overline{X}}_i^{\prime }{\overline{X}}_i\right)}^{-1}{\overline{X}}_i\right],$

And

(31) $C=\frac{{(\pi )}^{\frac{(N_i-n_i)p}{2}}{\Gamma }_p(\frac{n_i-k}{2})|H{|}^{-\frac{p}{2}}}{{\Gamma }_p(\frac{N_i-k}{2})|S_{Y_i}{|}^{\frac{n_i-k}{2}}}$

is the normalizing constant.

The joint posterior density of parameters (say) θ = (β,Σ)′ for the observed sample units Y_i and

unobserved population units ${\overline{Y}}_i$ can be determined as

(32) $f(\beta ,\Sigma |Y_i,{\overline{Y}}_i)\infty |\Sigma {|}^{-\frac{N_i+p+1}{2}}|I_p+{\Sigma }^{-1}Q{|}^{-\frac{v+p+N_i-1}{2}}$

where

(33) $Q=(Y_i-X_i\beta {)}^{\prime }(Y_i-X_i\beta )+({\overline{Y}}_i-{\overline{X}}_i\beta {)}^{\prime }({\overline{Y}}_i-{\overline{X}}_i\beta ).$

Now by using the Markov Chain Monte Carlo simulation method to equation (32), we can obtain simulated copies of the micropopulation data for the i^th small area.

The main steps involved with this process of spatial microdata simulation are as follows:

1. obtain a suitable joint prior distribution of the event under research E_i, say housing stress in the population at i^th small area, that is p(E_i) for ∀_i;

2. find the conditional distribution of unobserved sampling units, given the observed data, that is $p\left(y_{ij}:j\in {\overline{s}}_i|y_{ij}:j\in s_i\right)for{\forall }_i$;

3. derive the posterior distribution using Bayes theorem, that is p (θ | s, X); E_i ⊆ θ, where θ is the vector of model parameters and X is an auxiliary information vector; and

4. get simulated copies of the entire population from this posterior distribution by the MCMC simulation technique.

The key feature of this new method is that it can simulate complete scenarios of the whole micro-population in a small area, which means it can produce more reliable small area estimates and their variance estimation. It is also able to create the statistical reliability measures (for example, the Bayes credible region or confidence interval) of spatial microsimulation models’ estimates that are still unavailable in the literature.

However, in the new approach to find a suitable prior distribution for each interested event, as well as the appropriate model for linking between observed data and unobserved sampling units are difficult in practice.

This paper has briefly summarised the overall methods of small area estimation and explicitly described some methodological issues in the spatial microsimulation modelling arena. Review papers in small area estimation literature have regularly focused only methodologies on various statistical approaches including the area level and unit level modelling with E-BLUP, EB and HB methods. However, spatial microsimulation modelling has also been widely used in small area estimation, and recently classified as the geographic approach. Simulating a reliable synthetic spatial microdata is the key challenge in the indirect geographic approach of small area estimation. The review of different methodologies demonstrates that two reweighting methods – the GREGWT and CO are commonly used tools to produce small area microdata.

The GREGWT technique utilizes a truncated Chi-squared distance function and generates a set of new weights by minimising the total distance with respect to some constraint functions. The minimisation tool Lagrange multipliers has been used in the GREGWT process to minimise the distance function and it is based on the Newton-Raphson iterative process. Results show that sets of new weights can vary substantially with changing values of the vector of difference between the benchmark totals and sample based estimated totals. On the other hand, the combinatorial optimisation technique uses an intelligent searching algorithm simulated annealing – which selects an appropriate set of households’ from survey microdata that best fits to the benchmark constraints by minimising the total absolute error/distance with respect to the Metropolis Criterion. The new weights give the actual household units, which are the best representative combination. Thus, CO is a selection process to reach an appropriate combination of sample units rather than calibrating the sampling design weights to a set of new weights.

Findings reveal that the GREGWT and CO are using quite different iterative algorithms and their properties also vary, but their performances are fairly similar from the standpoint of use in SMM. The Chi-squared distance measures show more smooth fluctuations than the absolute distance measures. Besides, SMMs techniques are robust and have significant advantages. In particular, since the spatial microsimulation framework uses a list-based approach to microdata representation, it is possible to use the microdata file for further analysis and updating. Also, by linking spatial microsimulation models with static microsimulation models, it is possible to measure small area effects of policy changes.

Moreover, the study points out a new approach in the spatial microsimulation methodology. The new technique is based on the Bayesian prediction theory and can simulate complete scenarios of the whole population in each small area. As a result the process can yield more accurate and statistically reliable small area estimates compared to the estimates from the other reweighting techniques. Besides, the Bayesian prediction based microdata simulation is a probabilistic approach, which is quite different from the deterministic approach used in GREGWT and the intelligent searching tool simulated annealing used in CO. However, the new approach can use the generalised regression model operated in the GREGWT algorithm to link observed units in the sample and unobserved units in the population. In contrast, from the view point of the CO reweighting, it uses the MCMC simulation with a posterior density based iterative algorithm. Further account of the new approach and its practical applications on empirical data will be presented in our next manuscript. Future research may look into this option of methodological advancements by practicing it into all arenas of SMMs.

The Newton-Raphson iteration method is a root-finding algorithm for a nonlinear equation. The method is based on the first few terms of the Taylor series of a function. Let for a single variable nonlinear equation f(z) = 0, the Taylor series of f(z) about the point z=z₀+ε is expressed as

Where z₀ is an initial assumed root of f(z), f^′ represents the first order derivative and ε is a small arbitrary positive quantity. Keeping terms only to first order derivative, we have

Now (a2) is the equation of tangent line to the curve of f(z) at the point {z₀, f(z₀)}, and hence (0,z₀) is the interval where that tangent line intersects the horizontal axis at z₁ (Fig. a-1).

Figure a-1

Download asset Open asset

The expression in (a2) can be used to estimate the amount of adjustment for ε should require to converge to the accepted root starting from an initial assumed root value, z₀. From the relation in (a2), after setting f(z₀+ε)=0 and considering an arbitrary quantity ε=ε₀ we get

which is the first-order adjustment to the original root.

Now by considering z_i = z_i-1 + ε_i-1 for i = 1,2,…,r, we can subsequently obtain a new ε_i, for which

Let the process should be repeated until (r+1) times when a value of the arbitrary quantity, ε is reached to the accuracy level. In other words, the process should be repeated until (r+1) times when an estimated root of the function - (say) z_r+1, will converge to a precisely stable number or to an accepted root value. Hence the following algorithm can be applied iteratively to obtain an accepted root

The method uses this iterative equation in (a4) to approach one root of a function. A well-chosen initial root value can lead the convergence quickly (Fig. a-1). However the procedure can be unstable near a horizontal asymptote or a local extremum. Besides, this iteration method is easily adapted to deal with a set of equations for a function with vector variables when its second order derivative also exists.

Now equation (6) in GREGWT theory can be written as a function of the vector λ -

for j = 1,2,…,p;

where

$C=T_x-{\widehat{t}}_{x,s}$ is a known vector,

$d_k\{f^{-1}(x_k^{\prime }\lambda )-1\}$ is a scalar,

and the equation is nonlinear in the Lagrange multipliers vector, λ.

The equation (a5) can be solved by the above Newton-Raphson iterative procedure. Hence the iteration algorithm can be expressed as

where

λ_[r] is the value of the vector λ in the r^th iteration,

represents the Hessian matrix,

and

[l′(λ)]_[r]

defines the values of vector l’(λ), which are determined by the r^th iteration values of vector λ_[r].

Note that GREGWT stops iteration process when the condition

or when a predefined maximum iteration has been reached.

However, the ε_r can take any suitable positive arbitrary value and the choice is fully depending on our desired accuracy.

1. 1
   The 1998–99 Household Expenditure Survey, Australia: Confidentialised Unit Record Files (CURF), Technical Manual (2)

   1. ABS

   (2002)

   Canberra: Australian Bureau of Statistics.

   • Google Scholar
2. 2
   Statistical Matching of the HES and NHS: An Exploration of Issues in the use of Unconstrained and Constrained Approaches in Creating a Basefile for a Microsimulation Model of the Pharmaceutical Benefits Scheme

   1. ABS

   (2004)

   Canberra: Australian Bureau of Statistics.

   • Google Scholar
3. 3
   Statistical matching between different datasets: An application to the Spanish household survey (EPF90) and the income tax file (IRPF90)

   1. J Alegre
   2. J Arcarons
   3. S Calonge
   4. A Manresa

   (2000)
4. 4
   Creating small-area Income Estimates: spatial microsimulation modelling

   1. B Anderson

   (2007)
5. 5
   Exploring microsimulation methodologies for the estimation of household attributes

   1. D Ballas
   2. G Clarke
   3. I Turton

   (1999)

   paper presented at the, 4th International conference on GeoComputation, Virginia, USA, July 25–28.

   • Google Scholar
6. 6
   A spatial microsimulation approach to local labour market policy analysis

   1. D Ballas

   (2001)

   A spatial microsimulation approach to local labour market policy analysis, unpublished PhD thesis, School of Geography, University of Leeds, UK.

   • Google Scholar
7. 7
   A spatial microsimulation model for social policy evaluation

   1. D Ballas
   2. GP Clarke
   3. I Turton

   (2003)

   In: B Boots, R Thomas, editors. Modelling Geographical Systems, 70. Netherlands: Kluwer. pp. 143–168.

   • Google Scholar
8. 8
   Geography Matters: Simulating the local Impacts of National Social Policies

   1. D Ballas
   2. D Rossiter
   3. B Thomas
   4. GP Clarke
   5. D Dorling

   (2005)

   York: Joseph Rewntree Foundation.

   • Google Scholar
9. 9
   Modelling the socio-economic impacts of major job loss or gain at the local level: a spatial microsimulation framework

   1. D Ballas
   2. G Clarke
   3. j Dewhurst

   (2006)

   Spatial Economic Analysis 1:127–146.

   • Google Scholar
10. 10
    GREGWT and TABLE macros - User guide

    1. P Bell

    (2000)

    GREGWT and TABLE macros - User guide, ABS, Canberra, unpublished.

    • Google Scholar
11. 11
    Weighting and standard error estimation for ABS Household Surveys

    1. P Bell

    (2000a)

    Canberra: Australian Bureau of Statistics.

    • Google Scholar
12. 12
    SYNTHESIS- a synthetic spatial information system for urban and regional analysis: methods and examples

    1. M Birkin
    2. M Clarke

    (1988)

    Environment and Planning Analysis 20:1645–1671.

    • Google Scholar
13. 13
    Tools for Microeconomic Policy Analysis

    1. L Brown
    2. A Harding

    (2005)

    The new frontier of health and aged care: using microsimulation to assess policy options, Tools for Microeconomic Policy Analysis, Productivity Commission, Canberra.

    • Google Scholar
14. 14
    Modelling our future: population ageing, health and aged care

    1. SF Chin
    2. A Harding

    (2007)

    SpatialMSM, Modelling our future: population ageing, health and aged care, Amsterdam, North-Holland.

    • Google Scholar
15. 15
    Spatial microsimulation using synthetic small area estimates of income, tax and social security benefits

    1. SF Chin
    2. A. Harding
    3. R Lloyd
    4. J McNamara
    5. B Phillips
    6. Q.N. Vu

    (2005)

    Australasian Journal of Regional Studies 11:303–335.

    • Google Scholar
16. 16
    Regional Dimensions: Creating Synthetic Small-area Microdata and Spatial Microsimulation Models. Online Technical Paper - TP33

    1. SF Chin
    2. A Harding

    (2006)

    University of Canberra: NATSEM.

    • Google Scholar
17. 17
    Microsimulation methods in spatial analysis and planning

    1. M Clarke
    2. E. Holm

    (1987)

    Geografiska Annaler, Series B, Human Geography 69:145–164.

    • Google Scholar
18. 18
    The use of spatial microsimulation and geographic information systems (GIS) in benefit function transfer - an application to modelling the demand for recreational activities in Ireland

    1. J Cullinan
    2. S Hynes
    3. C O’Donoghue

    (2006)

    paper presented at the, 8th Nordic Seminar on Microsimulation models, Oslo, June 7–9.

    • Google Scholar
19. 19
    On a least squares adjustment of a sampled frequency table when the expected marginal totals are known

    1. WE Deming
    2. FF Stephan

    (1940)

    The Annals of Mathematical Statistics 11:427–444.

    • Google Scholar
20. 20
    A Model for Updating Census-Based Population and Household Information for Inter-Censal Years

    1. CJ Duley

    (1989)

    University of Leeds, UK: School of Geography.

    • Google Scholar
21. 21
    Subjective Bayesian models in sampling finite populations

    1. WA Ericson

    (1969)

    Journal of the Royal Statistical Society. Series B 31:195–233.

    • Google Scholar
22. 22
    A three dimensional furness procedure for calibarating gravity models

    1. SP Evans
    2. HR Kirby

    (1974)

    Transportation Research 8:105–122.

    • Google Scholar
23. 23
    An iterative procedure for estimation in contingency tables

    1. SE Fienberg

    (1970)

    The Annals of Mathematical Statistics 41:907–917.

    • Google Scholar
24. 24
    Small area estimation: an appraisal

    1. M Ghosh
    2. JNK Rao

    (1994)

    Statistical Science 9:55–93.

    • Google Scholar
25. 25
    Lifetime income distribution and redistribution: applications of a microsimulation model

    1. A Harding

    (1993)

    North-Holland: Amsterdam.

    • Google Scholar
26. 26
    Microsimulation and public policy, Contributions to economic analysis

    1. A Harding

    (1996)

    North-Holland: Amsterdam.

    • Google Scholar
27. 27
    Assessing poverty and Inequality at a detailed regional level: new advances in spatial microsimulation

    1. A Harding
    2. R. Lloyd
    3. A Bill
    4. A King

    (2003)

    In: M McGillivray, M Clarke, editors. Understanding Human Well-Being, 1. Helsinki: United Nation University Press. pp. 239–261.

    • Google Scholar
28. 28
    Modelling our future: population aging, social security and taxation, International symposia in economic theory and econometrics

    1. A Harding
    2. A Gupta

    (2007)

    Amsterdam: Elsevier.

    • Google Scholar
29. 29
    Improving work incentives and incomes for parents: the national and geographic impact of liberalising the family tax benefit income test

    1. A Harding
    2. QN Vu
    3. R Tanton
    4. Y. Vidyattama

    (2009)

    The Economic Record 85:48–58.

    • Google Scholar
30. 30
    Model-based small area estimation series no. 2: small area estimation project report

    1. P Heady
    2. P Clarke
    3. G Brown
    4. K. Ellis
    5. D Heasman
    6. S Hennell
    7. J Longhurst
    8. B Mitchell

    (2003)

    UK: Office for National Statistics.

    • Google Scholar
31. 31
    A Comparison of Synthetic Reconstruction and Combinatorial Optimisation Approaches to the Creation of Small-Area Microdata. Working Paper 2001/2

    1. Z Huang
    2. P Williamson

    (2001)

    University of Liverpool, UK: Population Microdata Unit, Department of Geography.

    • Google Scholar
32. 32
    Modelling our future: population ageing, health and aged care

    1. A King

    (2007)

    Providing income support services to a changing aged population in Australia: Centrelink’s Regional Microsimulation model, Modelling our future: population ageing, health and aged care, Amsterdam, North-Holland.

    • Google Scholar
33. 33
    Optimization by Simulated Annealing

    1. S Kirkpatrick
    2. CD Gelatt Jr.
    3. MP Vecchi

    (1983)

    Science 220:671–680.

    • Google Scholar
34. 34
    An objective Bayesian view of survey weights, O’Bayes 07

    1. R Little

    (2007)
35. 35
    Proceedings of the Survey Methods Section

    1. TP Liu
    2. MS Kovacevic

    (1997)

    An empirical study on categorically constrained matching, Proceedings of the Survey Methods Section, Canada, Statistical Society of Canada.

    • Google Scholar
36. 36
    Bayesian statistical inference for sampling a finite population

    1. AY Lo

    (1986)

    The Annals of Statistics 14:1226–1233.

    • Google Scholar
37. 37
    Development of CareMod/05. Online Technical Paper - TP32

    1. S Lymer
    2. L Brown
    3. A Harding
    4. M Yap
    5. SF Chin
    6. S Leicester

    (2006)

    University of Canberra: NATSEM.

    • Google Scholar
38. 38
    Regional disability estimates for New South Wales in 2001 using spatial microsimulation

    1. S Lymer
    2. L Brown
    3. M Yap
    4. A Harding

    (2008)

    Applied Spatial Analysis and Policy 1:99–116.

    • Google Scholar
39. 39
    A noninformative Bayesian approach to small area estimation

    1. G Meeden

    (2003)

    Survey Methodology 29:19–24.

    • Google Scholar
40. 40
    Microsimulation- a survey of principles, developments and applications

    1. J Merz

    (1991)

    International Journal of Forecasting 7:77–104.

    • Google Scholar
41. 41
    Equation of state calculations by fast computing machines

    1. N Metropolis
    2. AW Rosenbluth
    3. MN Rosenbluth
    4. AH Teller
    5. E Teller

    (1953)

    Journal of Chemical Physics 21:1087–1092.

    • Google Scholar
42. 42
    Statistical matching: A paradigm for assessing the uncertainty in the procedure

    1. C Moriarity
    2. F Scheuren

    (2001)

    Journal of Official Statistics 17:407–422.

    • Google Scholar
43. 43
    A note on Rubin’s statistical matching using file concatenation with adjusted weights and multiple imputations

    1. C Moriarity
    2. F Scheuren

    (2003)

    Journal of Business and Educational Studies, 21:65–73.

    • Google Scholar
44. 44
    WP 99/03

    1. P Norman

    (1999)

    Putting iterative proportional fitting on the researcher’s desk, WP 99/03, School of Geography, University of Leeds, UK.

    • Google Scholar
45. 45
    A new type of socio-economic system

    1. HG Orcutt

    (2007)

    116–123, Reprinted with permission in the, International Journal of Microsimulation, 1, Autumn, www.microsimulation.org/IJM, originally published in 1957 in, Review of Economics and Statistics, 39, 2.

    • Google Scholar
46. 46
    Small area estimation -new developments and directions

    1. D Pfeffermann

    (2002)

    International Statistical Review 70:125–143.

    • Google Scholar
47. 47
    Intelligent optimisation techniques: genetic algorithms, tabu search, simulated annealing and neural networks

    1. DT Pham
    2. D Karaboga

    (2000)

    London: Springer.

    • Google Scholar
48. 48
    A review of small area estimation problems and methodological developments. Online Discussion Paper -DP66

    1. A Rahman

    (2008a)

    University of Canberra: NATSEM.

    • Google Scholar
49. 49
    The possibility of using Bayesian prediction theory in small area estimation

    1. A Rahman

    (2008b)

    presentation to the ARCRNSISS/ANZRSAI Annual Conference.

    • Google Scholar
50. 50
    Bayesian Predictive Inference for Some Linear Models under Student-t Errors

    1. A Rahman

    (2008c)

    VDM Verlag: Saarbrucken.

    • Google Scholar
51. 51
    Some current trends in sample survey theory and methods (with discussion)

    1. JNK Rao

    (1999)

    Sankhya: The Indian Journal of Statistics; Series B 61:1–57.

    • Google Scholar
52. 52
    Small area estimation: update with appraisal

    1. JNK Rao

    (2002)

    In: N Balakrishnan, editors. Advances on Methodological and Applied Aspects of Probability and Statistics. New York: Taylor and Francis. pp. 113–139.

    • Google Scholar
53. 53
    Small Area Estimation

    1. JNK Rao

    (2003)

    New Jersey: John Wiley and Sons, Inc.

    • Google Scholar
54. 54
    Some new developments in small area estimation

    1. JNK Rao

    (2003a)

    JIRSS 2:145–169.

    • Google Scholar
55. 55
    Statistical Matching: A Frequentist Theory, Practical Applications, and Alternative Bayesian Approaches

    1. S Rassler

    (2002)

    Verlag: Springer.

    • Google Scholar
56. 56
    Data fusion: identification problems, validity, and multiple imputation

    1. S Rassler

    (2004)

    Austrian Journal of Statistics 33:153–171.

    • Google Scholar
57. 57
    An evaluation of statistical matching

    1. WL Rodgers

    (1984)

    Journal of Business and Economic Statistics 2:91–102.

    • Google Scholar
58. 58
    Combining sample and census data in small area estimates: iterative proportional fitting with standard software

    1. L Simpson
    2. M Tranmer

    (2005)

    The Professional Geographer 57:222–234.

    • Google Scholar
59. 59
    SPATIALMSM: The Australian spatial microsimulation model

    1. R Tanton

    (2007)

    The 1st General Conference of the International Microsimulation Association.

    • Google Scholar
60. 60
    Comparing two methods of reweighting a survey file to small area data - Generalised regression and Combinatorial optimisation

    1. R Tanton
    2. P. Williamson
    3. A Harding

    (2007)

    The 1st General Conference of the International Microsimulation Association.

    • Google Scholar
61. 61
    Housing unaffordability at the statistical local area level: new estimates using spatial microsimulation

    1. E Taylor
    2. A Harding
    3. R Lloyd
    4. M Blake

    (2004)

    Australian Journal of regional Studies 10:279–300.

    • Google Scholar
62. 62
    The case for small area microdata

    1. M Tranmer
    2. A Pickles
    3. E Fieldhouse
    4. M Elliot
    5. A Dale
    6. M. Brown
    7. D Martin
    8. D Steel
    9. C Gardiner

    (2005)

    Journal of the Royal Statistical Society: Series A 168:29–49.

    • Google Scholar
63. 63
    Simulated Annealing: Theory and Applications

    1. PJ van Laarhoven
    2. EH Aarts

    (1987)

    NY: Springer.

    • Google Scholar
64. 64
    An evaluation of the combinatorial optimisation approach to the creation of synthetic microdata

    1. D Voas
    2. P Williamson

    (2000)

    International Journal of Population Geography 6:349–366.

    • Google Scholar
65. 65
    Community Health Care Policies for the Elderly: A Microsimulation Approach

    1. P Williamson

    (1992)

    University of Leeds, UK: School of Geography.

    • Google Scholar
66. 66
    Microsimulation for Urban and Regional Policy Analysis

    1. P Williamson
    2. GP Clarke
    3. AT McDonald

    (1996)

    Estimating small area demands for water with the use of microsimulation, Microsimulation for Urban and Regional Policy Analysis, Pion, London.

    • Google Scholar
67. 67
    The estimation of population microdata by using data from small area statistics and sample of anonymised records

    1. P Williamson
    2. M Birkin
    3. P Rees

    (1998)

    Environment and Planning Analysis 30:785–816.

    • Google Scholar
68. 68
    CO Instruction Manual. Working Paper 2007/1

    1. P Williamson

    (2007)

    University of Liverpool, UK: Population Microdara Unit, Department of Geography.

    • Google Scholar
69. 69
    The Reliability of Using the Iterative Proportional Fitting Procedure

    1. DWS Wong

    (1992)

    The Professional Geographer 44:340.

    • Google Scholar

Author details

1. Azizur Rahman

   National Centre for Social and Economic Modelling (NATSEM), University of Canberra, Australia

   For correspondence

   Azizur.Rahman@natsem.canberra.edu.au

2. Ann Harding

   National Centre for Social and Economic Modelling (NATSEM), University of Canberra, Australia

3. Robert Tanton

   National Centre for Social and Economic Modelling (NATSEM), University of Canberra, Australia

4. Shuangzhe Liu

   National Centre for Social and Economic Modelling (NATSEM), University of Canberra, Australia