28 Mar
2006
28 Mar
'06
8:59 p.m.
Hi,
I was thinking about how to do exponential regression and came up with this
quick optimization. I was considering using it to fill in missing parts of a
Zipf distribution, but I have decided that the assumptions to do so are not
met, particularly considering that political manifestos in German (which has a
lot of compound words) have a pretty well-filled Zipf distribution whereas
manifestos in English (which has a lot of multi-word terms) do not. I am
curious as to how to adapt this estimator to time-series analysis and to make
it more robust.
Geoff
# the function to optimize
f <- function(par, X, Y, W=rep(1, nrow(X)))
{
beta <- par
beta[2] <- 1/par[2]
sum(t(W) %*% abs(exp(X %*% beta) - Y))
}
# read in the data and set up variables
table <- read.csv("table.csv", header=T)
Y <- as.matrix(cbind(table[7]))
X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
W <- as.matrix(table[6])
# do a linear regression on transformed Y values, taking the
# reciprocal of beta_2 for optimization simplicity
lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
b0 <- par
# extract a set of coefficients for initializing the optimization
betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
betahat <- betahat_
betahat[2] <- 1/betahat[2]
# plot
plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted price")
lines(X[,2], exp(X %*% betahat), col="blue")
b0[2] <- 1/b0[2]
lines(X[,2], exp(X %*% b0), col="red")
legend(x=0, y=120, legend=c("Naive Transformed Least Squares", "Absolute
Least
Error Predictor"), fill=c("red", "blue"), bty="n")
29 Mar
29 Mar
3:07 a.m.
Hi Geoff,
Exponential regression models are duration models (time until an event
happens) - so I'm not quite sure what you mean when you say you want to
adapt it to time-series analysis. Also zelig fits exponential models.
Best,
Ian
On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
Hi,
I was thinking about how to do exponential regression and came up with this
quick optimization. I was considering using it to fill in missing parts of a
Zipf distribution, but I have decided that the assumptions to do so are not
met, particularly considering that political manifestos in German (which has a
lot of compound words) have a pretty well-filled Zipf distribution whereas
manifestos in English (which has a lot of multi-word terms) do not. I am
curious as to how to adapt this estimator to time-series analysis and to make
it more robust.
Geoff
# the function to optimize
f <- function(par, X, Y, W=rep(1, nrow(X)))
{
beta <- par
beta[2] <- 1/par[2]
sum(t(W) %*% abs(exp(X %*% beta) - Y))
}
# read in the data and set up variables
table <- read.csv("table.csv", header=T)
Y <- as.matrix(cbind(table[7]))
X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
W <- as.matrix(table[6])
# do a linear regression on transformed Y values, taking the
# reciprocal of beta_2 for optimization simplicity
lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
b0 <- par
# extract a set of coefficients for initializing the optimization
betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
betahat <- betahat_
betahat[2] <- 1/betahat[2]
# plot
plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted price")
lines(X[,2], exp(X %*% betahat), col="blue")
b0[2] <- 1/b0[2]
lines(X[,2], exp(X %*% b0), col="red")
legend(x=0, y=120, legend=c("Naive Transformed Least Squares", "Absolute
Least
Error Predictor"), fill=c("red", "blue"), bty="n")
10:23 a.m.
New subject: [gov2001-l] Time Series
Zelig fits models of the form y=ae^(bx); I just reparameterized and fit the
model with another error function; the reparameterization below could be
improved in order to take advantage of greater floating point precision (which
Zelig presumeably does). Exponential models are also useful for modeling
population growth of yeast in vats and sundry other phenomena.
Something was mentioned in lecture while on the slide "Quantities of Interest"
about drawing from the log historical prices of stocks.
YourCast fits time series models.
http://gking.harvard.edu/yourcast/
Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
Hi Geoff,
Exponential regression models are duration models (time until an event
happens) - so I'm not quite sure what you mean when you say you want to
adapt it to time-series analysis. Also zelig fits exponential models.
Best,
Ian
On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
Hi,
I was thinking about how to do exponential regression and came up with this
quick optimization. I was considering using it to fill in missing parts of
a
Zipf distribution, but I have decided that the
assumptions to do so are not
met, particularly considering that political manifestos in German (which
has a
lot of compound words) have a pretty well-filled
Zipf distribution whereas
manifestos in English (which has a lot of multi-word terms) do not. I am
curious as to how to adapt this estimator to time-series analysis and to
make
it more robust.
Geoff
# the function to optimize
f <- function(par, X, Y, W=rep(1, nrow(X)))
{
beta <- par
beta[2] <- 1/par[2]
sum(t(W) %*% abs(exp(X %*% beta) - Y))
}
# read in the data and set up variables
table <- read.csv("table.csv", header=T)
Y <- as.matrix(cbind(table[7]))
X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
W <- as.matrix(table[6])
# do a linear regression on transformed Y values, taking the
# reciprocal of beta_2 for optimization simplicity
lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
b0 <- par
# extract a set of coefficients for initializing the optimization
betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
betahat <- betahat_
betahat[2] <- 1/betahat[2]
# plot
plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted
price")
lines(X[,2], exp(X %*% betahat),
col="blue")
b0[2] <- 1/b0[2]
lines(X[,2], exp(X %*% b0), col="red")
legend(x=0, y=120, legend=c("Naive Transformed Least Squares", "Absolute
Least
Error Predictor"), fill=c("red",
"blue"), bty="n")
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
7:32 p.m.
I took a closer look and decided that the whole thing is garbage. Still working
on it (spring break and replication pretty much done).
Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
Hi Geoff,
Exponential regression models are duration models (time until an event
happens) - so I'm not quite sure what you mean when you say you want to
adapt it to time-series analysis. Also zelig fits exponential models.
Best,
Ian
On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
Hi,
I was thinking about how to do exponential regression and came up with this
quick optimization. I was considering using it to fill in missing parts of
a
Zipf distribution, but I have decided that the
assumptions to do so are not
met, particularly considering that political manifestos in German (which
has a
lot of compound words) have a pretty well-filled
Zipf distribution whereas
manifestos in English (which has a lot of multi-word terms) do not. I am
curious as to how to adapt this estimator to time-series analysis and to
make
it more robust.
Geoff
# the function to optimize
f <- function(par, X, Y, W=rep(1, nrow(X)))
{
beta <- par
beta[2] <- 1/par[2]
sum(t(W) %*% abs(exp(X %*% beta) - Y))
}
# read in the data and set up variables
table <- read.csv("table.csv", header=T)
Y <- as.matrix(cbind(table[7]))
X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
W <- as.matrix(table[6])
# do a linear regression on transformed Y values, taking the
# reciprocal of beta_2 for optimization simplicity
lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
b0 <- par
# extract a set of coefficients for initializing the optimization
betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
betahat <- betahat_
betahat[2] <- 1/betahat[2]
# plot
plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted
price")
lines(X[,2], exp(X %*% betahat),
col="blue")
b0[2] <- 1/b0[2]
lines(X[,2], exp(X %*% b0), col="red")
legend(x=0, y=120, legend=c("Naive Transformed Least Squares", "Absolute
Least
Error Predictor"), fill=c("red",
"blue"), bty="n")
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
30 Mar
30 Mar
2:17 a.m.
I would be interested in learning how to adapt this sort of estimator to handle
serially correlated errors.
table <- read.csv("table.csv")
table <- table[ rev( c(1:nrow(table))),]
money <- 1000
commission = 8
ndays <- 250
starts <- 1:(nrow(table) - ndays)
ends <- starts + ndays
interest <- (table[ends, 7] - table[starts, 7]) / table[starts, 7]
contribution <- money * interest - 2 * commission
expected_contribution <- mean(contribution)
hist(contribution)
abline(v=expected_contribution, col="red")
print("expected_contribution:")
print(expected_contribution)
Quoting ghumphr(a)fas.harvard.edu:
I took a closer look and decided that the whole thing
is garbage. Still
working
on it (spring break and replication pretty much done).
Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
whereas
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi Geoff,
Exponential regression models are duration models (time until an event
happens) - so I'm not quite sure what you mean when you say you want to
adapt it to time-series analysis. Also zelig fits exponential models.
Best,
Ian
On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
> Hi,
>
> I was thinking about how to do exponential regression and came up with
this
> quick optimization. I was considering using
it to fill in missing parts
of
a
> Zipf distribution, but I have decided that the assumptions to do so are
not
met,
particularly considering that political manifestos in German (which
has a
> lot of compound words) have a pretty well-filled Zipf distribution
manifestos in English (which has a lot of multi-word terms) do not. I am
curious as to how to adapt this estimator to time-series analysis and to
make
it more robust.
Geoff
# the function to optimize
f <- function(par, X, Y, W=rep(1, nrow(X)))
{
beta <- par
beta[2] <- 1/par[2]
sum(t(W) %*% abs(exp(X %*% beta) - Y))
}
# read in the data and set up variables
table <- read.csv("table.csv", header=T)
Y <- as.matrix(cbind(table[7]))
X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
W <- as.matrix(table[6])
# do a linear regression on transformed Y values, taking the
# reciprocal of beta_2 for optimization simplicity
lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
b0 <- par
# extract a set of coefficients for initializing the optimization
betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
betahat <- betahat_
betahat[2] <- 1/betahat[2]
# plot
plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted
price")
lines(X[,2], exp(X %*% betahat),
col="blue")
b0[2] <- 1/b0[2]
lines(X[,2], exp(X %*% b0), col="red")
legend(x=0, y=120, legend=c("Naive Transformed Least Squares", "Absolute
Least
Error Predictor"), fill=c("red",
"blue"), bty="n")
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
11:44 a.m.
One last question on this subject (and a few words of praise for Zelig, an
excellent package)- can exponential models be fit reliably to multiple
independent variables?
I hope that all is well with you.
Quoting ghumphr(a)fas.harvard.edu:
I took a closer look and decided that the whole thing
is garbage. Still
working
on it (spring break and replication pretty much done).
Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
whereas
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi Geoff,
Exponential regression models are duration models (time until an event
happens) - so I'm not quite sure what you mean when you say you want to
adapt it to time-series analysis. Also zelig fits exponential models.
Best,
Ian
On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
> Hi,
>
> I was thinking about how to do exponential regression and came up with
this
> quick optimization. I was considering using
it to fill in missing parts
of
a
> Zipf distribution, but I have decided that the assumptions to do so are
not
met,
particularly considering that political manifestos in German (which
has a
> lot of compound words) have a pretty well-filled Zipf distribution
manifestos in English (which has a lot of multi-word terms) do not. I am
curious as to how to adapt this estimator to time-series analysis and to
make
it more robust.
Geoff
# the function to optimize
f <- function(par, X, Y, W=rep(1, nrow(X)))
{
beta <- par
beta[2] <- 1/par[2]
sum(t(W) %*% abs(exp(X %*% beta) - Y))
}
# read in the data and set up variables
table <- read.csv("table.csv", header=T)
Y <- as.matrix(cbind(table[7]))
X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
W <- as.matrix(table[6])
# do a linear regression on transformed Y values, taking the
# reciprocal of beta_2 for optimization simplicity
lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
b0 <- par
# extract a set of coefficients for initializing the optimization
betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
betahat <- betahat_
betahat[2] <- 1/betahat[2]
# plot
plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted
price")
lines(X[,2], exp(X %*% betahat),
col="blue")
b0[2] <- 1/b0[2]
lines(X[,2], exp(X %*% b0), col="red")
legend(x=0, y=120, legend=c("Naive Transformed Least Squares", "Absolute
Least
Error Predictor"), fill=c("red",
"blue"), bty="n")
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
11:54 a.m.
Hi Geoff,
Yes, exponential models can fit multiple independent variables - like any
other sysetmatic component.
Best,
Ian
On Thu, 30 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
One last question on this subject (and a few words of
praise for Zelig, an
excellent package)- can exponential models be fit reliably to multiple
independent variables?
I hope that all is well with you.
Quoting ghumphr(a)fas.harvard.edu:
I took a closer look and decided that the whole
thing is garbage. Still
working
on it (spring break and replication pretty much done).
Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
whereas
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Hi Geoff,
Exponential regression models are duration models (time until an event
happens) - so I'm not quite sure what you mean when you say you want to
adapt it to time-series analysis. Also zelig fits exponential models.
Best,
Ian
On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
> Hi,
>
> I was thinking about how to do exponential regression and came up with
this
> quick optimization. I was considering using
it to fill in missing parts
of
a
> Zipf distribution, but I have decided that the assumptions to do so are
not
met,
particularly considering that political manifestos in German (which
has a
> lot of compound words) have a pretty well-filled Zipf distribution
manifestos in English (which has a lot of multi-word terms) do not. I am
curious as to how to adapt this estimator to time-series analysis and to
make
it more robust.
Geoff
# the function to optimize
f <- function(par, X, Y, W=rep(1, nrow(X)))
{
beta <- par
beta[2] <- 1/par[2]
sum(t(W) %*% abs(exp(X %*% beta) - Y))
}
# read in the data and set up variables
table <- read.csv("table.csv", header=T)
Y <- as.matrix(cbind(table[7]))
X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
W <- as.matrix(table[6])
# do a linear regression on transformed Y values, taking the
# reciprocal of beta_2 for optimization simplicity
lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
b0 <- par
# extract a set of coefficients for initializing the optimization
betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
betahat <- betahat_
betahat[2] <- 1/betahat[2]
# plot
plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted
price")
lines(X[,2], exp(X %*% betahat),
col="blue")
b0[2] <- 1/b0[2]
lines(X[,2], exp(X %*% b0), col="red")
legend(x=0, y=120, legend=c("Naive Transformed Least Squares", "Absolute
Least
Error Predictor"), fill=c("red",
"blue"), bty="n")
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
1:31 p.m.
New subject: [gov2001-l] Alpha (Heterogeneity Coef.)
Hello All,
After extensive sleuthing, we still have a mystery coefficient in one of the
less crucial tables in our paper.
The table includes three models of a negative binomial regression with the
size of discussion networks as the independent variable. The explanatory
variables are (1) whether the respondent came from a 1985 or 2004 wave of
the survey, (2) the interviewers' perceptions of the respondent's
cooperativeness (cooperative, restless, hostile, with friendly/interested as
the omitted base case), (3) the number of the previous 10 questions the
respondent had refused to answer, and (4) demographic characteristics of the
respondents (education, sex, age, marriage status, race). Finally, each
model includes the following coefficient, labeled only as "Alpha
(Heterogeneity Coef.)," followed by an F-test.
We can't find anything in the article on what this might be. It doesn't
seem like it would be appropriate to use a Cronbach's alpha here (and the
coefficients are really low - 0.059 to 0.159, with the values getting lower
as additional explanatory variables of the above added to the model). And
if they are using Blau's heterogeneity index, we can't figure out how.
Does this jog anyone's memory? If not, we'll contact the authors, but I
wanted to try this last ditch effort.
Many thanks,
Abby
-----Original Message-----
From: gov2001-l-bounces(a)lists.fas.harvard.edu
[mailto:gov2001-l-bounces@lists.fas.harvard.edu]On Behalf Of Ian Brett Yohai
Sent: Thursday, March 30, 2006 4:54 PM
To: gov2001-l(a)lists.fas.harvard.edu
Subject: Re: [gov2001-l] Quantities of Interest
Hi Geoff,
Yes, exponential models can fit multiple independent variables - like any
other sysetmatic component.
Best,
Ian
On Thu, 30 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
One last question on this subject (and a few words of
praise for Zelig, an
excellent package)- can exponential models be fit reliably to multiple
independent variables?
I hope that all is well with you.
Quoting ghumphr(a)fas.harvard.edu:
> I took a closer look and decided that the whole thing is garbage. Still
> working
> on it (spring break and replication pretty much done).
>
> Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
>
> > Hi Geoff,
> >
> > Exponential regression models are duration models (time until an event
> > happens) - so I'm not quite sure what you mean when you say you want
to
> > adapt it to time-series analysis. Also
zelig fits exponential models.
> >
> > Best,
> > Ian
> >
> > On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
> >
> > > Hi,
> > >
> > > I was thinking about how to do exponential regression and came up
with
> this
> > > quick optimization. I was considering using it to fill in missing
parts
> of
> > a
> > > Zipf distribution, but I have decided that the assumptions to do so
are
> not
> > > met, particularly considering that political manifestos in German
(which
> > has a
> > > lot of compound words) have a pretty well-filled Zipf distribution
> whereas
> > > manifestos in English (which has a lot of multi-word terms) do not.
I am
> > > curious as to how to adapt this
estimator to time-series analysis
and to
> > make
> > > it more robust.
> > >
> > > Geoff
> > >
> > >
> > > # the function to optimize
> > > f <- function(par, X, Y, W=rep(1, nrow(X)))
> > > {
> > > beta <- par
> > > beta[2] <- 1/par[2]
> > > sum(t(W) %*% abs(exp(X %*% beta) - Y))
> > > }
> > >
> > > # read in the data and set up variables
> > > table <- read.csv("table.csv", header=T)
> > > Y <- as.matrix(cbind(table[7]))
> > > X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
> > > W <- as.matrix(table[6])
> > >
> > > # do a linear regression on transformed Y values, taking the
> > > # reciprocal of beta_2 for optimization simplicity
> > > lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
> > > par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
> > > b0 <- par
> > >
> > > # extract a set of coefficients for initializing the optimization
> > > betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
> > > betahat <- betahat_
> > > betahat[2] <- 1/betahat[2]
> > >
> > > # plot
> > > plot(X[,2], Y, main="Price of IBM vs Time",
xlab="day",
ylab="adjusted
> > price")
> > > lines(X[,2], exp(X %*% betahat), col="blue")
> > > b0[2] <- 1/b0[2]
> > > lines(X[,2], exp(X %*% b0), col="red")
> > > legend(x=0, y=120, legend=c("Naive Transformed Least Squares",
"Absolute
Least
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Error Predictor"), fill=c("red",
"blue"), bty="n")
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
1:51 p.m.
New subject: [gov2001-l] Alpha (Heterogeneity Coef.)
Hi,
If I remember correctly, when you run a negative binomial model in stata, it
includes some sort of hypothesis test to determine whether or not there is
enough overdispersion to conclude that the negative binomial model is an
appropriate alternative to the poisson. Is it possible that this is what
you're seeing? If you have access to stata, you might try running the
regression (the command is "nbreg") to see if it looks like that's what
the
author is reporting.
Good luck,
Dennis
-----Original Message-----
From: gov2001-l-bounces(a)lists.fas.harvard.edu
[mailto:gov2001-l-bounces@lists.fas.harvard.edu] On Behalf Of Abby
Williamson
Sent: Thursday, March 30, 2006 6:31 PM
To: gov2001-l(a)lists.fas.harvard.edu
Subject: [gov2001-l] Alpha (Heterogeneity Coef.)
Hello All,
After extensive sleuthing, we still have a mystery coefficient in one of the
less crucial tables in our paper.
The table includes three models of a negative binomial regression with the
size of discussion networks as the independent variable. The explanatory
variables are (1) whether the respondent came from a 1985 or 2004 wave of
the survey, (2) the interviewers' perceptions of the respondent's
cooperativeness (cooperative, restless, hostile, with friendly/interested as
the omitted base case), (3) the number of the previous 10 questions the
respondent had refused to answer, and (4) demographic characteristics of the
respondents (education, sex, age, marriage status, race). Finally, each
model includes the following coefficient, labeled only as "Alpha
(Heterogeneity Coef.)," followed by an F-test.
We can't find anything in the article on what this might be. It doesn't
seem like it would be appropriate to use a Cronbach's alpha here (and the
coefficients are really low - 0.059 to 0.159, with the values getting lower
as additional explanatory variables of the above added to the model). And
if they are using Blau's heterogeneity index, we can't figure out how.
Does this jog anyone's memory? If not, we'll contact the authors, but I
wanted to try this last ditch effort.
Many thanks,
Abby
-----Original Message-----
From: gov2001-l-bounces(a)lists.fas.harvard.edu
[mailto:gov2001-l-bounces@lists.fas.harvard.edu]On Behalf Of Ian Brett Yohai
Sent: Thursday, March 30, 2006 4:54 PM
To: gov2001-l(a)lists.fas.harvard.edu
Subject: Re: [gov2001-l] Quantities of Interest
Hi Geoff,
Yes, exponential models can fit multiple independent variables - like any
other sysetmatic component.
Best,
Ian
On Thu, 30 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
One last question on this subject (and a few words of
praise for
Zelig, an excellent package)- can exponential models be fit reliably
to multiple independent variables?
I hope that all is well with you.
Quoting ghumphr(a)fas.harvard.edu:
> I took a closer look and decided that the whole thing is garbage.
> Still working on it (spring break and replication pretty much done).
>
> Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
>
> > Hi Geoff,
> >
> > Exponential regression models are duration models (time until an
> > event
> > happens) - so I'm not quite sure what you mean when you say you
> > want
to
> > adapt it to time-series analysis. Also
zelig fits exponential models.
> >
> > Best,
> > Ian
> >
> > On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
> >
> > > Hi,
> > >
> > > I was thinking about how to do exponential regression and came
> > > up
with
> this
> > > quick optimization. I was considering using it to fill in
> > > missing
parts
> of
> > a
> > > Zipf distribution, but I have decided that the assumptions to do
> > > so
are
> not
> > > met, particularly considering that political manifestos in
> > > German
(which
> > has a
> > > lot of compound words) have a pretty well-filled Zipf
> > > distribution
> whereas
> > > manifestos in English (which has a lot of multi-word terms) do not.
I am
> > > curious as to how to adapt this
estimator to time-series
> > > analysis
and to
> > make
> > > it more robust.
> > >
> > > Geoff
> > >
> > >
> > > # the function to optimize
> > > f <- function(par, X, Y, W=rep(1, nrow(X))) {
> > > beta <- par
> > > beta[2] <- 1/par[2]
> > > sum(t(W) %*% abs(exp(X %*% beta) - Y)) }
> > >
> > > # read in the data and set up variables table <-
> > > read.csv("table.csv", header=T) Y <-
as.matrix(cbind(table[7]))
> > > X <- as.matrix(cbind(1, rev(c(1:nrow(Y))))) W <-
> > > as.matrix(table[6])
> > >
> > > # do a linear regression on transformed Y values, taking the #
> > > reciprocal of beta_2 for optimization simplicity lm.out <-
> > > lm(log(Y)~(X[,2]), weights=as.vector(W)) par <-
> > > c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2]) b0 <- par
> > >
> > > # extract a set of coefficients for initializing the
> > > optimization betahat_ <- optim(par, f, method="CG", X=X, Y=Y,
> > > W=W)$par betahat <- betahat_ betahat[2] <- 1/betahat[2]
> > >
> > > # plot
> > > plot(X[,2], Y, main="Price of IBM vs Time",
xlab="day",
ylab="adjusted
> > price")
> > > lines(X[,2], exp(X %*% betahat), col="blue") b0[2] <- 1/b0[2]
> > > lines(X[,2], exp(X %*% b0), col="red") legend(x=0, y=120,
> > > legend=c("Naive Transformed Least Squares",
"Absolute
Least
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
Error Predictor"), fill=c("red",
"blue"), bty="n")
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
5:30 p.m.
New subject: [gov2001-l] Zelig
Hi all,
does anyone know whether it is possible
to trick Zelig into computing quantities
of interest for models it doesn't
officially know about, e.g., a random
effects logit model?
Holger
ghumphr(a)fas.harvard.edu wrote:
One last question on this subject (and a few words of
praise for Zelig, an
excellent package)- can exponential models be fit reliably to multiple
independent variables?
I hope that all is well with you.
Quoting ghumphr(a)fas.harvard.edu:
--
Holger Lutz Kern
Graduate Student
Department of Government
Cornell University
Institute for Quantitative Social Science
Harvard University
1737 Cambridge Street N350
Cambridge, MA 02138
www.people.cornell.edu/pages/hlk23
I took a closer look and decided that the whole
thing is garbage. Still
working
on it (spring break and replication pretty much done).
Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
whereas
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l Hi Geoff,
Exponential regression models are duration models (time until an event
happens) - so I'm not quite sure what you mean when you say you want to
adapt it to time-series analysis. Also zelig fits exponential models.
Best,
Ian
On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
> Hi,
>
> I was thinking about how to do exponential regression and came up with
this
> quick optimization. I was considering using
it to fill in missing parts
of
a
> Zipf distribution, but I have decided that the assumptions to do so are
not
met,
particularly considering that political manifestos in German (which
has a
> lot of compound words) have a pretty well-filled Zipf distribution
manifestos in English (which has a lot of multi-word terms) do not. I am
curious as to how to adapt this estimator to time-series analysis and to
make
it more robust.
Geoff
# the function to optimize
f <- function(par, X, Y, W=rep(1, nrow(X)))
{
beta <- par
beta[2] <- 1/par[2]
sum(t(W) %*% abs(exp(X %*% beta) - Y))
}
# read in the data and set up variables
table <- read.csv("table.csv", header=T)
Y <- as.matrix(cbind(table[7]))
X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
W <- as.matrix(table[6])
# do a linear regression on transformed Y values, taking the
# reciprocal of beta_2 for optimization simplicity
lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
b0 <- par
# extract a set of coefficients for initializing the optimization
betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
betahat <- betahat_
betahat[2] <- 1/betahat[2]
# plot
plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted
price")
lines(X[,2], exp(X %*% betahat),
col="blue")
b0[2] <- 1/b0[2]
lines(X[,2], exp(X %*% b0), col="red")
legend(x=0, y=120, legend=c("Naive Transformed Least Squares", "Absolute
Least
Error Predictor"), fill=c("red",
"blue"), bty="n")
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
5:35 p.m.
New subject: [gov2001-l] Zelig
see the 'advanced' chapter in Zelig. what it requires is 'wrapping' a
function that does random effects logit, tho, which is probalby not the
answer you wanted!
Gary
On Thu, 30 Mar 2006, Holger Lutz Kern wrote:
Hi all,
does anyone know whether it is possible to trick Zelig into computing
quantities of interest for models it doesn't officially know about, e.g., a
random effects logit model?
Holger
ghumphr(a)fas.harvard.edu wrote:
One last question on this subject (and a few
words of praise for Zelig, an
excellent package)- can exponential models be fit reliably to multiple
independent variables?
I hope that all is well with you.
Quoting ghumphr(a)fas.harvard.edu:
I took a closer look and decided that the whole
thing is garbage. Still
working
on it (spring break and replication pretty much done).
Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l Hi Geoff,
Exponential regression models are duration models (time until an event
happens) - so I'm not quite sure what you mean when you say you want to
adapt it to time-series analysis. Also zelig fits exponential models.
Best,
Ian
On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
> Hi,
>
> I was thinking about how to do exponential regression and came up with
this
> quick optimization. I was considering using
it to fill in missing parts
of
a
> Zipf distribution, but I have decided that the assumptions to do so are
not
> met, particularly considering that political
manifestos in German (which
has a
> lot of compound words) have a pretty well-filled Zipf distribution
whereas
> manifestos in English (which has a lot of
multi-word terms) do not. I
> am
> curious as to how to adapt this estimator to time-series analysis and to
make
> it more robust.
>
> Geoff
>
>
> # the function to optimize
> f <- function(par, X, Y, W=rep(1, nrow(X)))
> {
> beta <- par
> beta[2] <- 1/par[2]
> sum(t(W) %*% abs(exp(X %*% beta) - Y))
> }
>
> # read in the data and set up variables
> table <- read.csv("table.csv", header=T)
> Y <- as.matrix(cbind(table[7]))
> X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
> W <- as.matrix(table[6])
>
> # do a linear regression on transformed Y values, taking the
> # reciprocal of beta_2 for optimization simplicity
> lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
> par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
> b0 <- par
>
> # extract a set of coefficients for initializing the optimization
> betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
> betahat <- betahat_
> betahat[2] <- 1/betahat[2]
>
> # plot
> plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted
price")
> lines(X[,2], exp(X %*% betahat), col="blue")
> b0[2] <- 1/b0[2]
> lines(X[,2], exp(X %*% b0), col="red")
> legend(x=0, y=120, legend=c("Naive Transformed Least Squares",
"Absolute
Least
> Error Predictor"), fill=c("red", "blue"),
bty="n")
>
>
>
>
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
31 Mar
31 Mar
12:36 a.m.
New subject: [gov2001-l] Zelig
Hi Holger,
I was thinking more about your problem, and I think what you want is the
function glmmPQL() in the package MASS.
it estimates generalized linear mixed effects models by repeated calls
to lme in
the nlme package. it suppports all the glm families, i.e., link=logit.
Hope this helps,
Delia
Quoting Gary King <king(a)harvard.edu>du>:
see the 'advanced' chapter in Zelig. what it requires is 'wrapping'
a function that does random effects logit, tho, which is probalby not
the answer you wanted!
Gary
On Thu, 30 Mar 2006, Holger Lutz Kern wrote:
Hi all,
does anyone know whether it is possible to trick Zelig into
computing quantities of interest for models it doesn't officially
know about, e.g., a random effects logit model?
Holger
ghumphr(a)fas.harvard.edu wrote:
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
One last question on this subject (and a few
words of praise for Zelig, an
excellent package)- can exponential models be fit reliably to multiple
independent variables?
I hope that all is well with you.
Quoting ghumphr(a)fas.harvard.edu:
I took a closer look and decided that the whole
thing is garbage. Still
working
on it (spring break and replication pretty much done).
Quoting Ian Brett Yohai <yohai(a)fas.harvard.edu>du>:
> Hi Geoff,
>
> Exponential regression models are duration models (time until an event
> happens) - so I'm not quite sure what you mean when you say you want to
> adapt it to time-series analysis. Also zelig fits exponential models.
>
> Best,
> Ian
>
> On Wed, 29 Mar 2006 ghumphr(a)fas.harvard.edu wrote:
>
>> Hi,
>>
>> I was thinking about how to do exponential regression and came up with
this
>> quick optimization. I was considering using it to fill in missing parts
of
> a
>> Zipf distribution, but I have decided that the assumptions to do so are
not
>> met, particularly considering that political manifestos in German (which
> has a
>> lot of compound words) have a pretty well-filled Zipf distribution
whereas
>> manifestos in English (which has a lot of multi-word terms) do
>> not. I am
>> curious as to how to adapt this estimator to time-series analysis and to
> make
>> it more robust.
>>
>> Geoff
>>
>>
>> # the function to optimize
>> f <- function(par, X, Y, W=rep(1, nrow(X)))
>> {
>> beta <- par
>> beta[2] <- 1/par[2]
>> sum(t(W) %*% abs(exp(X %*% beta) - Y))
>> }
>>
>> # read in the data and set up variables
>> table <- read.csv("table.csv", header=T)
>> Y <- as.matrix(cbind(table[7]))
>> X <- as.matrix(cbind(1, rev(c(1:nrow(Y)))))
>> W <- as.matrix(table[6])
>>
>> # do a linear regression on transformed Y values, taking the
>> # reciprocal of beta_2 for optimization simplicity
>> lm.out <- lm(log(Y)~(X[,2]), weights=as.vector(W))
>> par <- c(coefficients(lm.out)[1], 1/coefficients(lm.out)[2])
>> b0 <- par
>>
>> # extract a set of coefficients for initializing the optimization
>> betahat_ <- optim(par, f, method="CG", X=X, Y=Y, W=W)$par
>> betahat <- betahat_
>> betahat[2] <- 1/betahat[2]
>>
>> # plot
>> plot(X[,2], Y, main="Price of IBM vs Time", xlab="day",
ylab="adjusted
> price")
>> lines(X[,2], exp(X %*% betahat), col="blue")
>> b0[2] <- 1/b0[2]
>> lines(X[,2], exp(X %*% b0), col="red")
>> legend(x=0, y=120, legend=c("Naive Transformed Least Squares",
"Absolute
> Least
>> Error Predictor"), fill=c("red", "blue"),
bty="n")
>>
>>
>>
>>
> _______________________________________________
> gov2001-l mailing list
> gov2001-l(a)lists.fas.harvard.edu
> http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
>
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
6608
days inactive
6610
days old
gov2001@lists.gking.harvard.edu
11 comments
7 participants
participants (7)
-
abby.williamson＠gmail.com -
delia＠hss.caltech.edu -
dennis_feehan＠harvard.edu -
ghumphr＠fas.harvard.edu -
hlk23＠cornell.edu -
king＠harvard.edu -
yohai＠fas.harvard.edu