Hello everyone.
I'm doing logistic regression analysis of determinants of child stunting and would like to present the data using bar charts.
I've already done it but I would like to improve the layout of the charts.
The code that I use is:
I would like to:
1) Get rid of unnecessary space (e.g. no values below 1, hence don't need these values)
2) Make the graph vertical
3) Any other suggestion to improve the layout would be great
ALSO, for each OR graph I'm using separate logistic model, so the results are not adjusted. How to provide aOR bar chart for only one determinant? I tried to use aOR but ended up with all the determinants in one graph...
One more question about logistic model. I'm using SVY to analyze the data. Is it possible to provide smaller CI?
I'm also facing the problem of bigger number of insignificant results in aOR model...while OR model presents more significant results.
e.g. unadjusted
adjusted
Beforehand thank you!
I'm doing logistic regression analysis of determinants of child stunting and would like to present the data using bar charts.
I've already done it but I would like to improve the layout of the charts.
The code that I use is:
Code:
parmest, norestore eform egen axis=axis(estimate), label (parm) twoway bar estimate c_age, base(1) horizontal barw(.5) xline(1) ylab(, valuelabel) || rcap min95 max95 c_age, horizontal legend(order(1 "point estimate" 2 "95% conf. int.") pos(6)) xtitle("odds ratio") ytitle("child age in months")
1) Get rid of unnecessary space (e.g. no values below 1, hence don't need these values)
2) Make the graph vertical
3) Any other suggestion to improve the layout would be great
ALSO, for each OR graph I'm using separate logistic model, so the results are not adjusted. How to provide aOR bar chart for only one determinant? I tried to use aOR but ended up with all the determinants in one graph...
One more question about logistic model. I'm using SVY to analyze the data. Is it possible to provide smaller CI?
I'm also facing the problem of bigger number of insignificant results in aOR model...while OR model presents more significant results.
e.g. unadjusted
Code:
svy: logistic stunted i.c_age, baselevels svy: logistic stunted i.c_sex, baselevels
Code:
svy: logistic stunted i.c_age i.c_sex i.cbord ib(freq).mbmic ib(3).m_edu ib(1).initiation_bf ib(3).f_edu ib(5).wealth i.residence i.region, baselevels