Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Nick Cox <njcoxstata@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: nl hockey estimation |

Date |
Fri, 10 Aug 2012 18:28:52 +0100 |

You are quite right in terms of understanding the definition from scratch. My point was that Jordan's example had the right-hand part translated up by 100 relative to the left-hand part, and my code was a fix of that. Nick On Fri, Aug 10, 2012 at 6:18 PM, Steve Samuels <sjsamuels@gmail.com> wrote: > > Quite right, Nick. I was confused by the "y" on the RHS > of your y2 equation. > > I meant to write > gen y = cond(x<50, x, 100 -x) > > which _is_ a segmented line. > > I'd just note that your y2 is perhaps easier to > understand as: > > gen y2 = cond(x<50, x, 3*x-100) > > > Al, I gave the earliest reference I knew to the "hockey stick" terminology > problem in: > > http://www.stata.com/statalist/archive/2010-04/msg01712.html > > > nlhockey.ado is part of the loghockey package. One can see this and others that > Mark Lunt has written by typing: > "net from http://personalpages.manchester.ac.uk/staff/mark.lunt/"; > > Steve > sjsamuels@gmail.com > > On Aug 10, 2012, at 10:49 AM, Nick Cox wrote: > > No; on this occasion I meant what I wrote. What you suggest shows > another discontinuity. > > Nick > > On Fri, Aug 10, 2012 at 3:42 PM, Steve Samuels <sjsamuels@gmail.com> wrote: >> >> Nick, I think you meant: >> >> gen y2 = cond(x <50, y, 100 - y) >> >> Steve >> sjsamuels@gmail.com >> >> On Aug 9, 2012, at 9:09 PM, Nick Cox wrote: >> >> The word "clearly" here is questionable. Your test data show a big >> discontinuity; they aren't a segmented line which is what the model is >> looking for. The least squares criterion is being used and -nl- does >> the best it can to minimise the sum of _squared_ errors. The built-in >> aversion to very large errors is what is biting here. >> >> If you work instead with >> >> gen y2 = cond(x < 50, y, y - 100) >> nl hockey y2 x >> >> you will get what you expect. >> >> On this evidence the program is fine, but your test example won't work >> as you expect under LS. At a wild guess, L1-norm might give something >> nearer splitting the difference. >> >> Nick >> >> On Fri, Aug 10, 2012 at 12:34 AM, Jordan Silberman >> <silberman.stata@gmail.com> wrote: >> >>> I'm attempting to identify a breakpoint in a regression using the -nl >>> hockey- command (described here: >>> http://personalpages.manchester.ac.uk/staff/mark.lunt/nlhockey.hlp). >>> >>> When I test this command using simple simulated data, I find that the >>> command doesn't identify the correct breakpoint. Here's an example: >>> >>> set obs 100 >>> gen x = _n >>> gen y = x if x < 50 >>> replace y = x*3 if x > 49 >>> nl hockey y x >>> >>> The breakpoint should clearly be at 50; however, command output >>> identifies the breakpoint at 32.7. >>> >>> So, 2 questions: >>> >>> 1. Why might the -nl hockey- command be computing the wrong breakpoint? >>> >>> 2. Can anyone recommend an alternate approach to identifying the >>> breakpoint in a 2-piece regression? Best would be something that's >>> been implemented in Stata in a straightforward way. * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**References**:**st: nl hockey estimation***From:*Jordan Silberman <silberman.stata@gmail.com>

**Re: st: nl hockey estimation***From:*Nick Cox <njcoxstata@gmail.com>

**Re: st: nl hockey estimation***From:*Steve Samuels <sjsamuels@gmail.com>

**Re: st: nl hockey estimation***From:*Nick Cox <njcoxstata@gmail.com>

**Re: st: nl hockey estimation***From:*Steve Samuels <sjsamuels@gmail.com>

- Prev by Date:
**Re: Re: st: Regression with different firms** - Next by Date:
**st: RE: dash line in histogram** - Previous by thread:
**Re: st: nl hockey estimation** - Next by thread:
**st: saving and using chi2 distances from correspondence analysis** - Index(es):