STATA 5 hypothesis testing-1

docx

School

Indiana University, Bloomington *

*We aren’t endorsed by this school

Course

MISC

Subject

Economics

Date

Feb 20, 2024

Type

docx

Pages

4

Uploaded by EarlNightingale4120

Report
STATA Handout #5 – Adi Sarangee. 9:45 Tu/Th Econometrics HYPOTHESIS TESTING NOTE: Please submit your STATA handout via Canvas. Also, please print a copy and bring to class on the due date as we will be using this STATA output in our class discussion. NOTE: Please read the data dictionary before doing the analysis. It is very important that you are familiar with the variables in the data set to fully understand the analysis. You want to use the EXCEL data file 2018 state data will be used for this assignment. HYPOTHESIS TESTING This Stata assignment involves learning how to conduct hypothesis testing for MCLRMs. Example #1: Descriptive Statistics Find the descriptive statistics for the data set Commands/Code: tabstat cdr spending obese alcohol cig2 ins ed2 picap98 urban minority, statistics (n mean median sd max min cv skewness) correlate
Example #2: Estimate Model You want to estimate the effect of spending, obese, alcohol, cig2, ins, ed2, picap98, urban, minority on cdr and test a couple hypotheses using a t-test: 1) obesity has no effect on the crude death rate, 2) the effect of a one unit increase in the amount of alcohol consumed per capita on the crude death rate has the same effect of a one unit increase in the amount of cigarette consumption per capita. Commands/Code: regress cdr spending obese alcohol cig2 ins ed2 picap98 urban minority lincom alcohol-cig2 NOTES 1. The regress command reports the t-statistic for the zero null hypothesis.
2. The lincom command reports the t-statistic for the zero null hypothesis that the difference between the coefficients of alcohol and cig2 is zero, i.e. equal effects. Example #3: Estimate Model Next, you want to test a few additional hypotheses using a F-test: 1) obesity, alcohol consumption, and cigarette consumption do not have a joint effect on the crude death rate, 2) the effect of consuming one additional gallon of alcohol is equal to the effect of consuming 20 additional packs of cigarettes on crude death rate, 3) the effect of education is equal to the effect of spending and the effect of having insurance is equal to the effect of per capital income, and 4) the effect of percentage living in urban area has same effect of percentage of minority population Commands/Code: regress cdr spending obese alcohol cig2 ins ed2 picap98 urban minority test (obese=0) (alcohol=0) (cig2=0) test alcohol=20*cig2 test (ed2=spending) (ins=picap98) test (urban=minority)
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
NOTES 1. The test command reports the F-statistic for one or more linear hypotheses. When testing two or more hypotheses, each hypothesis must be placed in parentheses.