Parameter Estimation and Evaluation Empirical Studies and Statistical Inference Empirical Studies and Statistical Inference Various data imperfections may arise in practice,such as: multicolinearity or near-multicolinearity; conditional heteroskekasticity and autocorrelation in error terms; censored data; truncated data; contaiminated data; errors in variables; -missing observations; endogeneity; curse of dimensionality; weak instruments; unobserved counterfactuals; partial identification,and etc. Big Data,Machine Learning and Statistics Introduction to Statistics and Econometrics Juy8,2020 11/70
Parameter Estimation and Evaluation Big Data, Machine Learning and Statistics Introduction to Statistics and Econometrics July 8, 2020 11/70 Empirical Studies and Statistical Inference Empirical Studies and Statistical Inference
Parameter Estimation and Evaluation Empirical Studies and Statistical Inference Empirical Studies and Statistical Inference Refined statistical inference procedures are developed. Key assumptions for statistical modelling and inference: -Stochasticity.The system under study is stochastic or random in nature. Model Uniqueness:The probability law of the DGP can be uniquely charac- terized by a mathematical model. -Correct Model Specification.The unique model is assumed to be correctly specified. -Sampling Inference for Population.It has been a basic approach in statistics to use a sample or a data set,which contains a subset of information of the DGP,to make inference of the population,namely the probability law of the DGP. Representative Sampling.There is no sample selection bias,and the observed data usually has a moderately sample size. Statistical Significance.Whether an explanatory variable or covariate is impor- tant is judged by its statistical inference,typically at a prespecified (e.g.5%) significance level,or equivalently using a P-value of the test statistic Big Data,Machine Learning and Statistics Introduction to Statistics and Econometrics July8,2020 1270
Parameter Estimation and Evaluation Big Data, Machine Learning and Statistics Introduction to Statistics and Econometrics July 8, 2020 12/70 Empirical Studies and Statistical Inference Empirical Studies and Statistical Inference
CONTENTS 10.1 Introduction 10.2 Empirical Studies and Statistical Inference 10.3 Important Features of Big Data 10.4 Big Data Analysis and Statistics 10.5 Machine Learning and Statistics 10.6 Conclusion Big Data,Machine Learning and Statistics Introduction to Statistics and Econometrics July8,2020 13/70
Big Data, Machine Learning and Statistics Introduction to Statistics and Econometrics July 8, 2020 13/70 10.1 Introduction 10.2 Empirical Studies and Statistical Inference 10.3 Important Features of Big Data 10.4 Big Data Analysis and Statistics 10.5 Machine Learning and Statistics 10.6 Conclusion CONTENTS
Parameter Estimation and Evaluation Important Features of Big Data Important Features of Big Data Big Data arises due to the application of information technolo- gies,particularly internet and mobile internet technologies. The exponential growth of internet-connected devices and sen- sors is the major contributor to the massive data and storage. An example of Big data and its use is the development of autonomous vehicles. -In economics and business,massive Big Data arise due to the widespread computer-based business transactions and activities thought internets and mobile internets. In China,the number of internet and mobile internet users has reached more than 829 millions by the end of 2018, which is more than the total number of the users of internet and mobile internets in U.S.and European Union. Big Data,Machine Learning and Statistics Introduction to Statistics and Econometrics July8,2020 14/70
Parameter Estimation and Evaluation Big Data, Machine Learning and Statistics Introduction to Statistics and Econometrics July 8, 2020 14/70 Important Features of Big Data Important Features of Big Data
Parameter Estimation and Evaluation Important Features of Big Data Important Features of Big Data Big data comes from a variety of sources,including: computer-based business transactions; scanning machines record all tick-by-tick transactions in financial markets and credit card purchases in supermarkets and online orderings. information from social media and websites; various websites of business organizations and government sec- tors; -weibo,facebook. Big Data,Machine Learning and Statistics Introduction to Statistics and Econometrics Juy8,2020 15/70
Parameter Estimation and Evaluation Big Data, Machine Learning and Statistics Introduction to Statistics and Econometrics July 8, 2020 15/70 Important Features of Big Data Important Features of Big Data