Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

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

From |
<S.Jenkins@lse.ac.uk> |

To |
<statalist@hsphsun2.harvard.edu> |

Subject |
st: RE: SE/CI for difference in transition matrix row proportions after -svy tabulate- twoway |

Date |
Wed, 4 Apr 2012 11:08:52 +0100 |

------------------------------ Date: Tue, 3 Apr 2012 17:32:11 +0000 From: "Scholes, Shaun" <s.scholes@ucl.ac.uk> Subject: st: RE: SE/CI for difference in transition matrix row proportions after -svy tabulate- twoway I'm having an early evening junior moment (and about to log off) but could you not: svy: mean bu_SA, over(Lbu_SA) and use lincom to estimate the difference, and obtain SE/CIs for the difference using that? Best wishes Shaun ------------------------------------ Oh to have junior moments! Shaun: thanks, your suggestion provides a solution for me. Here's an illustration using the union data set: . webuse union, clear (NLS Women 14-24 in 1968) . svyset idcode pweight: <none> VCE: linearized Single unit: missing Strata 1: <one> SU 1: idcode FPC 1: <zero> . ge byte Lunion = L.union (16639 missing values generated) . svy: tabulate Lunion union, row se (running tabulate on estimation sample) Number of strata = 1 Number of obs = 9561 Number of PSUs = 3621 Population size = 9561 Design df = 3620 ------------------------------------- | 1 if union Lunion | 0 1 Total ----------+-------------------------- 0 | .9203 .0797 1 | (.0033) (.0033) | 1 | .2645 .7355 1 | (.0109) (.0109) | Total | .7741 .2259 1 | (.0064) (.0064) ------------------------------------- Key: row proportions (linearized standard errors of row proportions) Pearson: Uncorrected chi2(1) = 4073.6377 Design-based F(1, 3620) = 3197.8295 P = 0.0000 . svy: mean union, over(Lunion) (running mean on estimation sample) Survey: Mean estimation Number of strata = 1 Number of obs = 9561 Number of PSUs = 3621 Population size = 9561 Design df = 3620 0: Lunion = 0 1: Lunion = 1 -------------------------------------------------------------- | Linearized Over | Mean Std. Err. [95% Conf. Interval] -------------+------------------------------------------------ union | 0 | .0796877 .0033145 .0731892 .0861862 1 | .7354597 .010891 .7141066 .7568127 -------------------------------------------------------------- . lincom [union]1 - [union]0 ( 1) - [union]0 + [union]1 = 0 ------------------------------------------------------------------------ ------ Mean | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------- ------ (1) | .655772 .0115293 56.88 0.000 .6331673 .6783766 ------------------------------------------------------------------------ ------ . ret list scalars: r(df) = 3620 r(se) = .0115293178519787 r(estimate) = .6557719519645262 - -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of S.Jenkins@lse.ac.uk Sent: 03 April 2012 18:13 To: statalist@hsphsun2.harvard.edu Subject: st: SE/CI for difference in transition matrix row proportions after -svy tabulate- twoway I'm having an early evening senior moment, and can't figure out how to calculate, from saved results after -svy tabulate-, the difference between two elements of a 2x2 transition matrix and the associated SE/CI. I've been browsing svy help and documentation, and can't find the answer directly. Part of my problem is not understanding precisely what is stored in the saved variance-covariance matrix. Below is example output from a simplified example. I have a binary measure of receipt at t-1 and at t, and cross-tabulate them. The two transition proportions of interest are the entry rate, P(0,1), which is estimated to be 0.0335 in the example, and the stayer rate, P(1,1), which is estimated to be 0.6553 in the example. I want not only the the difference P(1,1) - P(0,1), but also a SE/CI for the difference, which I was assuming I could calculate from the saved results. Suggestions please. . svy: tabulate Lbu_SA bu_SA, row se (running tabulate on estimation sample) Number of strata = 1 Number of obs = 75988 Number of PSUs = 9036 Population size = 75988 Design df = 9035 - ------------------------------------------- 1:R's BU | receives | IS|UBIS|U | B|JSA, |1:R's BU receives IS|UBIS|UB|JSA t-1 | 0 1 Total - ----------+-------------------------------- 0 | .9665 .0335 1 | (8.2e-04) (8.2e-04) | 1 | .3447 .6553 1 | (.0095) (.0095) | Total | .9255 .0745 1 | (.0023) (.0023) - ------------------------------------------- Key: row proportions (linearized standard errors of row proportions) Pearson: Uncorrected chi2(1) = 2.62e+04 Design-based F(1, 9035) = 1.47e+04 P = 0.0000 . mat list e(b) e(b)[1,4] p11 p12 p21 p22 y1 .96649524 .03350476 .34470377 .65529623 . mat list e(V_row) symmetric e(V_row)[2,2] r1: r2: r1 r1 r1:r1 4.394e-06 r2:r1 -4.394e-06 4.394e-06 . mat list e(V) symmetric e(V)[4,4] p11 p12 p21 p22 p11 6.751e-07 p12 -6.751e-07 6.751e-07 p21 1.477e-07 -1.477e-07 .00008959 p22 -1.477e-07 1.477e-07 -.00008959 .00008959 Stephen ------------------ Professor Stephen P. Jenkins <s.jenkins@lse.ac.uk> Department of Social Policy and STICERD London School of Economics and Political Science Houghton Street, London WC2A 2AE, UK Tel: +44(0)20 7955 6527 Changing Fortunes: Income Mobility and Poverty Dynamics in Britain, OUP 2011, http://ukcatalogue.oup.com/product/9780199226436.do Survival Analysis Using Stata: http://www.iser.essex.ac.uk/survival-analysis Downloadable papers and software: http://ideas.repec.org/e/pje7.html Please access the attached hyperlink for an important electronic communications disclaimer: http://lse.ac.uk/emailDisclaimer * * 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/

- Prev by Date:
**Re: st: Different coefficients for xtivreg2** - Next by Date:
**st: save coef** - Previous by thread:
**st: Discrete time with shared frailty** - Next by thread:
**st: save coef** - Index(es):