首页 >> 学术论文 >> 驾驶行为分析

Smallest Acceptable Sample Size Estimation in Coefficient Estimation and Accuracy Prediction

发表时间:   阅读次数:

ABSTRACT: Determination of adequate sample size is critical in experimental design. Adequate sample size should be specified so researchers can make accurate inferences about their studied populations. However, the number of samples typically collected is largely subject to the expense of data collection. Working out the methodology of estimating the required number of subjects based on an initially small number is a more proper way for researchers to determine the necessary sample size in the experiment. Accordingly, this study estimates the smallest acceptable sample size, with emphasis on the accuracy of parameter coefficient estimation and accuracy of prediction made from regression equations (APCEAP), that is needed to work with a relatively small dataset for selected significant variables. This methodology is flexible and scalable, which is subject to users applications to all other experimental situations.
A relatively small sample size of 10 drivers was collected through stratified sampling to infer a theoretically adequate sample size. Using the APCEAP procedure in a freeway design safety evaluation, 30 drivers was determined to be the necessary sample size. To validate the appropriateness of this procedure, more than sufficient sample of 50 drivers was recruited. The smallest acceptable sample size was determined backwardly, based on the parameter coefficient convergence trends of the mean squared error (MSE) curves of significant variables. The clear converging trends of the MSE curves indicated that 30 was an acceptable sample size.

Bowen Cai, Xuesong Wang, Xiaohan Yang. Smallest Acceptable Sample Size Estimation in Coefficient Estimation and Accuracy Prediction. Transportation Research Board 99th Annual Meeting, Washington D.C., USA, 2020. 1.12-16.  

©CopyRight 2003-2012   同济大学交通运输工程学院

备案号:沪ICP备13005359号-1