window push log using  FramSBPbmiMulti.log, replace
window push *  FramSBPbmiMulti.log
window push *
window push *  Framingham data set: Multiple regression analysis of effect of bmi on sbp
window push *
window push set more on
window push use "2.20.Framingham.dta", clear
window push regress sbp bmi
window push scatter sbp  bmi, msymbol(Oh)						///
window push    || lfit sbp bmi, ytitle(Systolic Blood Pressure)			
window push more
window push regress sbp age
window push scatter sbp  age, msymbol(Oh)						///
window push    || lfit sbp age, ytitle(Systolic Blood Pressure)			
window push more
window push regress sbp scl
window push scatter sbp  scl, msymbol(Oh)						///
window push    || lfit sbp scl, ytitle(Systolic Blood Pressure)			
window push more
window push graph matrix sbp bmi age scl if month==1 & sex==2   , msymbol(oh)
window push more
window push *
window push *  Use multiple regression models with interaction terms to analyze
window push *  the effects of sbp, bmi, age and scl on sbp.
window push *
window push generate woman = sex -1
window push label define truth 0 "False" 1 "True"
window push label values woman truth
window push generate agewoman = age*woman
window push generate bmiwoman = bmi*woman
window push generate sclwoman = scl*woman
window push regress sbp bmi age scl woman bmiwoman agewoman sclwoman
window push regress sbp bmi age scl woman bmiwoman agewoman 
window push regress sbp bmi age scl woman  agewoman
window push *
window push *  Fit a model of sbp against bmi age scl and sex with
window push *  interaction terms.  The variables woman, bmiwoman,
window push *  agewoman, and sclwoman have been previously defined.
window push *
window push stepwise, pe(.1): regress sbp bmi  age scl woman bmiwoman agewoman sclwoman
window push more
window push stepwise, pr(.1): regress sbp bmi  age scl woman bmiwoman agewoman sclwoman  
window push more
window push predict yhat, xb
window push predict res, rstudent
window push lowess res yhat,  bwidth(0.2) msymbol(oh) color(gs10) lwidth(thick)     ///
window push     yline(-1.96 0 1.96) ylabel(-2 (2) 6) ytick(-2 (1) 6) 		///
window push     xlabel(100 (20) 180) xtitle(Expected SBP)
window push more
window push *
window push *  Illustrate influence of individual data points on
window push *  the parameter estimates of linear regression.
window push *
window push drop res
window push keep if id > 2000 & id <= 2050
window push regress sbp bmi age scl woman agewoman, level(50)
window push predict res, rstudent
window push predict cook, cooksd
window push label variable res "Studentized Residual"
window push label variable cook "Cook's Distance"
window push scatter cook res, ylabel(0 (.1) .5) xlabel(-2 (1) 5)
window push list cook res id bmi sbp if res > 2
window push regress  sbp bmi age scl woman agewoman if id ~= 2049, level(50)
window push log close