A continuous form of post-stratification

1979 ◽  
Vol 31 (2) ◽  
pp. 271-277
Author(s):  
David R. Brillinger
Keyword(s):  
Continuous Form ◽  
Post Stratification

Realized and potential efficiency for post-stratified estimation in a national forest inventory

2021 ◽  
Author(s):  
James A. Westfall ◽  
Andrew J. Lister ◽  
John W. Coulston ◽  
Ronald E. McRoberts
Keyword(s):  
Remotely Sensed ◽  
National Forest Inventory ◽  
Data Types ◽  
Remotely Sensed Data ◽  
Large Area ◽  
Forest Inventories ◽  
Temporal Misalignment ◽  
Total Tree ◽  
Potential Efficiency ◽  
Post Stratification

Post-stratification is often used to increase the precision of estimates arising from large-area forest inventories with plots established at permanent locations. Remotely sensed data and associated spatial products are often used for developing the post-stratification, which offers a mechanism to increase precision for less cost than increasing the sample size. While important variance reductions have been shown from post-stratification, it remains unknown where observed gains lie along the continuum of possible gains. This information is needed to determine whether efforts to further improve post-stratification outcomes are warranted. In this study, two types of ‘optimal’ post-stratification were compared to typical production-based post-stratifications to estimate the magnitude of remaining gains possible. Although the ‘optimal’ post-stratifications were derived using methods inappropriate for operational usage, the results indicated that substantial further increases in precision for estimates of both forest area and total tree biomass could be obtained with better post-stratifications. The potential gains differed by the attribute being estimated, the population being studied, and the number of strata. Practitioners seeking to optimize post-stratification face challenges such as evaluation of numerous auxiliary data sources, temporal misalignment between plot observations and remotely sensed data acquisition, and spatial misalignment between plot locations and remotely sensed data due to positional errors in both data types.

scholarly journals Is the ‘Local Candidate’ Advantage a Myth? Analysing the Effects of Localism in the 2015 UK General Election

2016 ◽  
Author(s):  
Javier Sajuria
Keyword(s):  
Mediation Analysis ◽  
General Election ◽  
Data Sources ◽  
Election Study ◽  
Multilevel Regression ◽  
Factual Information ◽  
Diverse Range ◽  
Local Candidate ◽  
Post Stratification

This paper studies the question of the so-called electoral advantage of local candidates. We use a diverse range of data sources to estimate whether a candidate residing in the same constituency they compete has any advantage at all. We then compare the effect of the factual information against a measure of perception of residence, taken from the British Election Study Internet Panel. We propose different methodological innovations from traditional analyses of this issue. We first concentrate on the top two candidates of the most competitive constituencies, and use a measurement of perception calculates using Multilevel Regression with Post-stratification. We use mediation analysis to estimate the overall effects. Our findings show that local candidates have an advantage only if they are perceived as local, and that incumbents are usually perceived as more local than challengers.

Measuring Wages Worldwide

2013 ◽  
pp. 100-119 ◽  
Author(s):  
Stephanie Steinmetz ◽  
Damian Raess ◽  
Kea Tijdens ◽  
Pablo de Pedraza
Keyword(s):  
Developing Countries ◽  
Data Collection ◽  
Demographic Variables ◽  
Web Survey ◽  
Data Collection Tool ◽  
Individual Level ◽  
Estimated Parameters ◽  
The Difference ◽  
The Web ◽  
Post Stratification

This chapter discusses the potentials and constraints of using a volunteer Web survey as a worldwide data collection tool for wages. It provides a detailed description of the bias related to individual-level wages and core socio-demographic and employment-related variables across selected developed and developing countries and evaluates the efficiency of post-stratification weights in adjusting these biases. The results confirm that Web samples are particularly attractive to younger persons, part-timers, and persons working in non-manual occupations. This can be observed across countries, although the strength of the bias differs between them. With respect to the efficiency of post-stratification weights, the results are inconclusive. Whereas it is advisable to implement weights for descriptive purposes of socio-demographic variables, the contrary holds in case of wages. Additionally, weights can have the opposite effect by (moderately) increasing the difference in the estimated parameters between the reference and the Web sample.

Post-stratification as a tool for enhancing the predictive power of classification methods

2021 ◽  
pp. 125-130
Author(s):  
Francesco D. d'Ovidio ◽  
Angela Maria D'Uggento ◽  
Rossana Mancarella ◽  
Ernesto Toma
Keyword(s):  
Logit Model ◽  
Predictive Power ◽  
Training Sample ◽  
Classification Problems ◽  
Target Variable ◽  
Predictive Capacity ◽  
Logit Transformation ◽  
Segmentation Analysis ◽  
Decision Making Model ◽  
Post Stratification

It is well known that, in classification problems, the predictive capacity of any decision-making model decreases rapidly with increasing asymmetry of the target variable (Sonquist et al., 1973; Fielding 1977). In particular, in segmentation analysis with a categorical target variable, very poor improvements of purity are obtained when the least represented modality counts less than 1/4 of the cases of the most represented modality. The same problem arises with other (theoretically more exhaustive) techniques such as Artificial Neural Networks. Actually, the optimal situation for classification analyses is the maximum uncertainty, that is, equidistribution of the target variable. Some classification techniques are more robust, by using, for example, the less sensitive logit transformation of the target variable (Fabbris & Martini 2002); however, also the logit transformation is strongly affected by the distributive asymmetry of the target variable. In this paper, starting from the results of a direct survey in which the target (binary) variable was extremely asymmetrical (10% vs. 90%, or greater asymmetry), we noted that also the logit model with the most significant parameters had very reduced fitting measures and almost zero predictive power. To solve this predictive issue, we tested post-stratification techniques, artificially symmetrizing a training sample. In this way, a substantially increase of fitting and predictive capacity was achieved, both in the symmetrized sample and, above all, in the original sample. In conclusion of the paper, an application of the same technique to a dataset of very different nature and size is described, demonstrating that the method is stable even in the case of analysis executed with all data of a population.

Sign in / Sign up

Export Citation Format

Share Document