I have a typical question if someone can help.
Construct two bivariate data sets (choose number of cases to suit yourself), each with correlation above 0.9, so that the combined data set has negative correlation.
5
-
1$\begingroup$ Welcome to Cross Validated! What progress have you made with this self-study question? $\endgroup$– DaveCommented Oct 12, 2022 at 23:08
-
$\begingroup$ Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. $\endgroup$– Community BotCommented Oct 13, 2022 at 3:06
-
$\begingroup$ See stats.stackexchange.com/a/366667/919, stats.stackexchange.com/q/478463/919, stats.stackexchange.com/q/316319/919 and more generally see our posts on Simpson's Paradox. $\endgroup$– whuber ♦Commented Oct 13, 2022 at 16:54
-
$\begingroup$ I am able to get two set of bivariate data with correlation greater than .9. But I don't know how to combine them to get a negative correlation $\endgroup$– puth puraCommented Oct 16, 2022 at 7:41
-
$\begingroup$ This excercise deals with a common misconception found among many beginning statisticians. If we merge two data sets each with high positive correlation, it seems reasonable to expect that the merged data will also have positive correlation. This exercise asks you to find a counter example. $\endgroup$– puth puraCommented Oct 16, 2022 at 10:11
Add a comment
|