gov1000-list December 2002

gov1000-list@lists.fas.harvard.edu

19 participants
83 discussions

by Yongwook Ryu

dear all, for some reason i can't get get into my vnc session through the secure crt. I was able to, just a moment ago, but not now. does anyone know why? yongwook ----------------------------- Yongwook Ryu PhD Candidate Department of Government Harvard University Tel:617-493-3397 Email: yryu(a)fas.harvard.edu -----------------------------

21 years, 4 months

VNC connection error

by Tiffany Chiemi Nagano

arrggg... is anyone else getting this error? Exception connecting to localhost:8: VNC authentication failed - too many tries I'm trying to work from my home computer because I cant get to the yard yet, but it's not working, even though I've done it before! I need to work on the problem set... help! ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Tiffany C. Nagano Harvard College '05 nagano(a)fas.harvard.edu 313 Mather House Mail Center (617) 493-7370

21 years, 4 months

vnc server

by dkane＠latte.harvard.edu

I think that other people had similar problems last night. Don't hesitate to bring this to the attention of the nice folks at HMDC. Dave Xiaojun Yan writes: > Hi, Dave and Tao > > I don't know what happened but my R session was kept being > killed--although I am not the one who use the most system resources. > Even, my VNC session (26) was totally killed last night so that I am > current starting from the beginning of the semester...... > > xiaojun > > _____________________________ > Xiaojun Yan > Department of Government > Littauer Center, North Yard > Harvard University > Cambridge, MA 02138 > Telephone: (617)869-6815 > www.people.fas.harvard.edu/~yan2 > > > -- David Kane Lecturer In Government 617-563-0122 dkane(a)latte.harvard.edu Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html

21 years, 4 months

Help on 2d

by Traci Burch

Okay, I've got to the point where I can get "step" to actually start with all 99 independent variables and take some variables away (one at a time) until it gets all the way down to the intercept. this is my command line: try <- lm(Y ~ ., data = x) step(try, scale=100) sparing you the details, I looked over the R output and ran regression on the ten variables it gave somewhere around "AIC = -77.09". I got an R^2 of .28. Questions: What's AIC? why does scale set the max AIC and then go backwards from there? Traci

21 years, 4 months

Q 2(a)

by Yongwook Ryu

Dear all, i can't figure out why i get this error message in Q2(a) > set.seed(123) > x<-as.data.frame(matrix(rnorm(100*100), 100, 100)) > names(x)[1]<-"Y" > gk<-cor(x$Y, x) Error in cor(x$Y, x) : unused argument(s) ( ...) please help.... yongwook ----------------------------- Yongwook Ryu PhD Candidate Department of Government Harvard University Tel:617-493-3397 Email: yryu(a)fas.harvard.edu -----------------------------

21 years, 4 months

Re: did you get step for 2e?

by Traci Burch

oh, and do you have NA's for the last ten vars when it stops? its b/c you're trying to estimate starting from 100 variables with only ten cases. throw out the cases with the least correlation (until you have 90 vars) and start from there. T ----- Original Message ----- From: "Olivia Lau" <olau(a)fas.harvard.edu> To: "Traci Renee Burch" <tburch(a)fas.harvard.edu> Sent: Wednesday, December 11, 2002 11:20 PM Subject: did you get step for 2e? > Traci, did you get step () to work for the new model we have to estimate > in 2e? My command only runs it once for all 100 variables, and it > was the same one feom 2d...help? > > Thanks. > > Olivia. >

21 years, 4 months

2d woes continued...

by Phillip Y. Lipscy

Dear All, So out of frustration at step/add1/drop1, etc., we wrote a function designed to give us a high R^2 through a pure looping process. Unfortunately, the highest thing we got after doing several iterations is lower than what you get by simply taking the V's with the highest correlations with Y (0.18). Please advise... In the meantime, I'm going to take the k term out of the loop and try to run it with r^2 = 0.19 overnight to see if anything shows up. -Phillip. --- findfunc <- function(){ r <- 0 k <- 0 while(r < 0.16 & k < 1000){ a <- sample(2:99, 10) sumry <- summary(lm(x$Y ~ x[,a[1]] + x[,a[2]] + x[,a[3]] + x[,a[4]] + x[,a[5]] + x[,a[6]] + x[,a[7]] + x[,a[8]] + x[,a[9]] + x[,a[10]])) r <- sumry$adj.r.squared k <- k + 1 } return(sumry) } > findfunc() Call: lm(formula = x$Y ~ x[, a[1]] + x[, a[2]] + x[, a[3]] + x[, a[4]] + x[, a[5]] + x[, a[6]] + x[, a[7]] + x[, a[8]] + x[, a[9]] + x[, a[10]]) Residuals: Min 1Q Median 3Q Max -1.9248 -0.7724 -0.1722 0.6625 2.4000 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 0.002306 0.111871 0.021 0.9836 x[, a[1]] 0.179260 0.111769 1.604 0.1123 x[, a[2]] -0.097365 0.112073 -0.869 0.3873 x[, a[3]] -0.127174 0.107072 -1.188 0.2381 x[, a[4]] 0.207749 0.109559 1.896 0.0612 . x[, a[5]] 0.257978 0.122177 2.112 0.0375 * x[, a[6]] 0.014120 0.118444 0.119 0.9054 x[, a[7]] -0.264335 0.115807 -2.283 0.0248 * x[, a[8]] -0.264225 0.110827 -2.384 0.0192 * x[, a[9]] -0.168252 0.103627 -1.624 0.1080 x[, a[10]] 0.042248 0.116425 0.363 0.7176 --- Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1 Residual standard error: 1.061 on 89 degrees of freedom Multiple R-Squared: 0.2547, Adjusted R-squared: 0.1709 F-statistic: 3.041 on 10 and 89 DF, p-value: 0.002338 ## ## Just taking the highest correlates ## > summary(lm(Y ~ V50 + V55 + V36 + V57 + V84 + V48 + V85 + V33 + V16 + V54, data = x)) Call: lm(formula = Y ~ V50 + V55 + V36 + V57 + V84 + V48 + V85 + V33 + V16 + V54, data = x) Residuals: Min 1Q Median 3Q Max -2.22011 -0.57443 -0.01890 0.57753 3.67152 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) -0.004837 0.107877 -0.045 0.9643 V50 0.286877 0.119519 2.400 0.0185 * V55 0.212099 0.108986 1.946 0.0548 . V36 0.210628 0.114766 1.835 0.0698 . V57 0.155421 0.099233 1.566 0.1208 V84 0.228806 0.106233 2.154 0.0340 * V48 0.168072 0.105212 1.597 0.1137 V85 0.136183 0.112276 1.213 0.2284 V33 0.166785 0.121525 1.372 0.1734 V16 0.183758 0.107128 1.715 0.0898 . V54 0.015593 0.112396 0.139 0.8900 --- Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1 Residual standard error: 1.053 on 89 degrees of freedom Multiple R-Squared: 0.2661, Adjusted R-squared: 0.1836 F-statistic: 3.226 on 10 and 89 DF, p-value: 0.001382 ------------------------------------------------- Phillip Y. Lipscy Perkins Hall Room #129 35 Oxford Street Cambridge, MA 02138 (617)493-4893 lipscy(a)fas.harvard.edu Ph.D. Candidate Harvard University, FAS, Department of Government -------------------------------------------------

21 years, 4 months

Re: [gov1000-list] everyone, pls check using ps to see whether ...

by Tao Li

it might be that the system kills everything and you need to log back. u need to go back to the beginning of the semester (as if) and get a vnc # first. this might be the case. your document is saved, though. Tao Li -------------- www.people.fas.harvard.edu/~li7 On Thu, 12 Dec 2002, Xiaojun Yan wrote: > Hi, Tao > > It seems I cann't have my VNC connected to the server. It always says > that "connection refused". What is the probable problem? > > Thanks. > > xiaojun > > _____________________________ > Xiaojun Yan > Department of Government > Littauer Center, North Yard > Harvard University > Cambridge, MA 02138 > Telephone: (617)869-6815 > www.people.fas.harvard.edu/~yan2 > > On Thu, 12 Dec 2002, Tao Li wrote: > > > > > u are actually the one who is causing problem. ^|^ > > thanks. > > > > Tao Li > > -------------- > > www.people.fas.harvard.edu/~li7 > > > > > > _______________________________________________ > > gov1000-list mailing list > > gov1000-list(a)fas.harvard.edu > > http://www.fas.harvard.edu/mailman/listinfo/gov1000-list > > >

21 years, 5 months

server frozen

by Tiffany Chiemi Nagano

it's FROZEN!!!! help!!! ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Tiffany C. Nagano Harvard College '05 nagano(a)fas.harvard.edu 313 Mather House Mail Center (617) 493-7370

21 years, 5 months

server is frozen!

by Olivia Lau

Hey, what gives? Not nice! Olivia.

21 years, 5 months

← Newer
1
2
3
4
5
6
7
8
9
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

gov1000-list December 2002