The Chinese Stata Conference took place on 16 August 2020 in cooperation with the Wuhan University School of Economics and Management.
Proceedings
8:45–9:45 | Fusion application of Stata and Python
Abstract:
Fusion application with Python is a new feature of Stata 16.
It allows Stata to run Python programs freely. I will
demonstrate the possibilities of expanding Stata and Python
through this series of examples.
Additional information: Hua Peng
StataCorp LLC
|
9:45–10:30 | Link Stata to Chinese maps
Abstract:
I will introduce the principal methodology with which Stata reads
the geographic information from Baidu Maps and Gaode Maps, as well
as some map commands we developed.
cngcode converts Chinese addresses to latitude and longitude.
cnaddress converts latitude and longitude to Chinese iconic
geographical locations. cntraveltime can calculate the traffic
distance between two geographical locations (you can even choose
different traffic modes). cnmapsearch can search for various
geographical keywords such as subway stations, hospitals, cafes,
barbecue stalls, etc., within a few kilometers of a given location.
These commands are convenient for empirical research on
geography and transportation and have broad application prospects
in finance, economics, and sociology.
Additional information: Li Chuntao
Zhongnan University of Economics and Law
|
10:40–12:00 | Using Stata 16's lasso features for prediction and inference
Abstract:
Lasso and elastic net are two popular machine-learning methods. In this
presentation, I discuss Stata 16's new lasso features for prediction and
inference.
I will demonstrate how lasso-type techniques can be used for prediction
with linear, binary, and count outcomes. I will then show why these methods
are effective and how they work. The increasing availability of high-dimensional
data and increasing interest in more realistic functional forms have sparked a
renewed interest in automated methods for selecting the covariates to include
in a model. I discuss the promises and perils of model selection and pay special
attention to some new estimators that provide reliable inference after model
selection.
Additional information: Di Liu
StataCorp
|
12:00–12:40 | Visualization and modeling method of epidemic data of COVID-19
Abstract:
Since COVID-19 was found in 2019, the global spread has attracted extensive
attention. I will summarize basic methods for visualization of epidemic data
and explore the temporal and spatial characteristics of the global spread of
COVID-19.
I will then introduce a modeling method for forecasting the epidemic trend of
COVID-19 so as to better adjust the policy of the epidemic prevention and
control of the situation. Finally, I will summarize the matters needing attention
in the epidemic data analysis of the situation.
Additional information: Xiao Guangen
Wuhan University
|