st: Trouble Producing a Regression Discontinuity Graph with a Specified Bandwidth

[Dara Shifrer <Dara.Shifrer@rice.edu>]

I am using the rd command 
(http://ideas.repec.org/c/boc/bocode/s456888.html) in Stata/MP 13.0, 
Windows. I am only able to obtain a discontinuity graph if I do not 
specify a bandwidth. In other words, once I use both the ‘graph’ and 
‘bw’ options, I receive an ‘invalid syntax’ error message. I wonder if 
anyone else is having this problem, or sees the error in my code? Thank 
you for your consideration, Dara Shifrer
. rd tka11tmegn tb10a2m tv10cgim if _mj==1 & tv10sbjtm==1 & tb11excac!=1 
& ta10elgb5!=0 & ta10elgb5!=2, bw(.75)
Three variables specified; jump in treatment

at Z=0 will be estimated. Local Wald Estimate

is the ratio of jump in outcome to jump in treatment.

Assignment variable Z is tv10cgim

Treatment variable X_T is tb10a2m

Outcome variable y is tka11tmegn

(197247 missing values generated)

(197247 missing values generated)

(197247 missing values generated)

Estimating for bandwidth .75

Estimating for bandwidth .375

Estimating for bandwidth 1.5

------------------------------------------------------------------------------

tka11tmegn |Coef.Std. Err.zP>|z|[95% Conf. Interval]

-------------+----------------------------------------------------------------

numer |.111445.12124410.920.358-.126189.3490791

denom |.7869252.08285899.500.000.6245247.9493256

lwald |.1416209.15575890.910.363-.163661.4469027

numer50 |.059701.16841680.350.723-.2703899.3897919

denom50 |.6870157.11258226.100.000.4663586.9076729

lwald50 |.086899.24565710.350.724-.3945801.5683781

numer200 |.0817475.08397430.970.330-.0828391.246334

denom200 |.8488443.052343916.220.000.7462522.9514364

lwald200 |.0963044.09941470.970.333-.0985449.2911537

------------------------------------------------------------------------------

. rd tka11tmegn tb10a2m tv10cgim if _mj==1 & tv10sbjtm==1 & tb11excac!=1 
& ta10elgb5!=0 & ta10elgb5!=2, bw(.75) graph
Three variables specified; jump in treatment

at Z=0 will be estimated. Local Wald Estimate

is the ratio of jump in outcome to jump in treatment.

Assignment variable Z is tv10cgim

Treatment variable X_T is tb10a2m

Outcome variable y is tka11tmegn

Command used for graph: lpoly; Kernel used: triangle (default)

invalid syntax

r(198);

.



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/