* Screening Your Data for Potential Errors * Pavel SolĂ­s * 180.334 (02) Econometrics * October 2019 * =========================================== sysuse auto describe destring // if a numeric variable was read as a string drop keep // careful * Examining data list varlist in 1/10 summarize varlist, detail codebook price mpg tabulate rep78 list varlist if condition count if condition correlate rep78-weight pwcorr rep78-weight pwcorr rep78-weight, sig obs * Graphical methods for inspecting data stem rep78 // stem-and-leaf plot scatter mpg weight scatter mpg weight, mlabel(make) twoway (scatter mpg weight, mlabel(make)) (lfit mpg weight) graph box // boxplot graph matrix price mpg weight // scatterplot matrix histogram price, normal bin(20) kdensity price, normal * Others: symplot, qnorm, pnorm * Transforming variables: ladder, gladder