Day | Topics | Statistics | General / Graphs |
1 | One sample from normal distribution |
ci var ttest var=hypothesis centile var, c(2.5 97.5) meansd |
log close log using file sum var graph box var histogram var, norm qnorm var |
2 |
Two samples from normal distribution Nonparametric test Samplesize |
ttest var, by(group) sdtest var, by(group) ranksum var, by(group) db sampsi |
dotplot var,over(group) For graphs: Naming, combining and saving |
3 |
Paried data Nonparametric test Samplesize |
ttest var1=var2 signrank var1=var2 db sampsi |
For graphs: More data in the same plot using twoway |
4 |
Binary data One and two samples Risk difference, risk ratio, odds ratio Chi-square test Paired Spearmans test Samplesize |
ci var,bin prtest var=hypothesis bitest var=hypothesis cs var group,or woolf cc var group tab2 var1 var2, chi2 exact mcc var1 var2 spearman var1 var2 db sampsi |
|
5 |
Linear regression Comparing two lines Analysis of covariance |
regres var var1 predict newvar1,xb predict newvar2,res regres var i.var1##var2 regres var i.var1 var2 lincom ..... pwcorr var1 var2 |
twoway /// (scatter var var1) /// (lfit var var1) /// (lfit var var1,stdf) |
6 |
One way analysis of variance Bartlett test Kruskal-Wallis test |
oneway var group,tabu regres var i.group adjust, by(group) ci kwallis var, by(group) |
|
7 |
Odds ratios Simple Logistic regression |
logit var var1 logit var var1,or logit var i.group logit var i.group##var1 lincom ....,or |
|
8 |
Survival data Kaplan-Meier curves Median survival Proportinal hazard (Cox) regression |
stset eventtime /// ,fail(status) /// enter(startdate) /// origin(startdate) /// scale(365.25) stci,p(25) by(group) stcox i.group,nohr stcox i.group |
st graph,gw st graph,by(group) stcoxkm graph,by(group) stphplot,by(group) nolntime |