Comparing median at 2 points

[Buyadaa Oyunchimeg]

Dear Stata users,

I am wondering, is it possible to compare the median of continuous variable (V1-slope) at 2 different points of V2 (compare median of slope at 8 and 10)? And what kind of statistical test can be applied for this analysis?



Thank you.
Best wishes

[Clyde Schechter]

Well, looking closely at your scatter plot, there are only a very few points with baseline viral load = 8 or = 10 exactly. So I doubt that doing this would be fruitful. But if you wanted to compare median slope with viral load close to 8 compared to viral load close to 10, where "close" is defined liberally enough to leave you with a decent number of data points, that might work out.

So you might try

gen byte bv_approx = 8 if inrange(baseline_viral_load, 7.5, 8.5)
replace bv_approx = 10 if inrange(baseline_viral_load, 9.5, 10.5)

tabstat slope, by(bv_approx) statistics(p50)
median slope, by(bv_approx)

Read -help median- for more information about the test of equality of medians.

[Buyadaa Oyunchimeg]

As always, thank you so much for your help prof.Schechter.

[Kudakwashe Chipunza]

Dear Stata users,

I am trying to run the following code:

set seed 1000

sqreg FV_index i.CQFIN i.age i.gender i.employment i.location i.CINCOME i.education i.marital Fin_literacy_index i.social_grants, q(.25 .50 .75 .95) reps(1000)

where: FV_index is an index of financial vulnerability and i.CQFIN are quintiles of financial inclusion index


After estimating the model, i have attempted to run the slope equality test for the quintiles of financial inclusion index (CQFIN) but it giving me an error.

Can one do slope equality test on quintiles?

[Leonardo Guizzetti]

Kudakwashe Chipunza , please start a new thread with your question as it has nothing to do with the original question here.