Skip to main content

Questions tagged [seemingly-unrelated-regressions]

SUR stands for "Seemingly Unrelated Regressions", an econometric technique for fitting several models (w/ different response variables) simultaneously. SUR may be more efficient than fitting the models separately.

Filter by
Sorted by
Tagged with
1 vote
0 answers
20 views

Large $N$, small $T$ in SUR: workaround using system GMM

Consider a system of linear equations as in seemingly unrelated regression (SUR). If the number of equations $N$ is large relative to the sample size $T$, the weighting matrix in SUR (i.e. the error ...
Richard Hardy's user avatar
2 votes
1 answer
65 views

Why does system GMM fail due to computationally singular system in my setup?

I am estimating a system of seemingly unrelated regressions (SUR) with gmm::sysGmm in R. Each of the equations has one unique regressor and one common regressor. ...
Richard Hardy's user avatar
2 votes
1 answer
101 views

System GMM yields identical results for any weighting matrix

I am estimating a system of seemingly unrelated regressions (SUR) in R. Each of the equations has one unique regressor and one common regressor. I am using ...
Richard Hardy's user avatar
2 votes
1 answer
56 views

Why does `systemfit` yield different results for OLS and WLS under cross-equation restrictions?

I am following up on the question "Why does systemfit yield identical results for OLS and WLS?". It deals with estimating a system of linear equations ...
Richard Hardy's user avatar
2 votes
1 answer
63 views

SUR estimated via `systemfit` vs `sysGmm`: different standard errors

I am estimating a system of seemingly unrelated regressions (SUR) in R. Each of the equations has one unique regressor and one common regressor. I have two alternative implementations: one via ...
Richard Hardy's user avatar
2 votes
2 answers
79 views

Why does `systemfit` yield identical results for OLS and WLS?

I am estimating a system of seemingly unrelated regressions (SUR) using the systemfit package in R. Each of the equations has one unique regressor and one common ...
Richard Hardy's user avatar
2 votes
0 answers
57 views

SUR estimated by OLS: restricted model has higher likelihood

I have three versions of a system of linear regression models: M, M0 and M00. I think each ...
Richard Hardy's user avatar
0 votes
0 answers
26 views

How should I deal with an ordered logit model with numerous, mutually exclusive dummy variables?

I an trying to estimate an ordered logit model where the DV is a likert-scale response (1-5) and I have 6 independent dummy variables representing whether an observation belongs to one of six mutually-...
Haris's user avatar
  • 21
0 votes
1 answer
247 views

Testing equality of coefficients from regression models with long format dataframes

How can I test the equality of coefficients from multiple regression models? I want to compare multiple measurement methods to see if they capture the dynamic processes of the same construct in ...
kleinemaus's user avatar
1 vote
0 answers
37 views

Can I estimate a panel VARX as a SUR?

If my panel vector autoregression with exogenous variables (VARX) is unrestricted, and the same variables appear on every equation. Can I estimate it as a panel SUR?
user392544's user avatar
1 vote
0 answers
58 views

Seemingly Unrelated regression with Truncated data

I have sales data (quantity sold (tons) and the prices that those forest products were sold ($/ton)) of three forest products (e.g., sawtimber, chip-and-saw and pulpwood) and sales of these products ...
kabindra shahi's user avatar
0 votes
1 answer
119 views

Implementing SUR on weighted regression models

I have two equations: fit1= svyglm(Y1 ~ X1 + X2 + X3, design= design.mnps, data= data) fit2= svyglm(Y2 ~ X1 + X2 + X3, design= design.mnps, data= data) ...
User123's user avatar
  • 11
1 vote
2 answers
92 views

Hayashi Econometrics Seemingly Unrelated Regressions (SUR) Eq 4.5.13'-15'

According to 4.5.13 $\hat{A}_{mh}=\left(\frac{1}{n}\sum_{i=1}^{n}z_{im}x_{i}^{\prime}\right)\left(\frac{1}{n}\sum_{i=1}^{n}x_{i}x_{i}^{\prime}\right)^{-1}\left(\frac{1}{n}\sum_{i=1}^{n}x_{i}z_{ih}^{\...
Hagan Ross's user avatar
1 vote
0 answers
44 views

When is seemingly unrelated regression equivalent to a confirmatory factor analysis/structural equation model?

One way of describing a seemingly unrelated regression model is: $y_n \sim \beta x_n + \epsilon_j$ where $\epsilon_n \sim MultivariateNormal(0, \Sigma)$. We could rewrite this $y_n \sim MVN(\beta ...
socialscientist's user avatar
1 vote
0 answers
101 views

SUR regression (which fails for too big in R) has identical variables, so is mostly multiple OLS; does error term effects matter?

I originally posted this to stackoverflow, and was advised by I_O that the question would be appreciated more here. I am not sure how to reformat this to be appropriate here, so I am copying it over ...
Andrew A's user avatar
  • 108
1 vote
0 answers
36 views

Appropriate term for "seemingly unrelated regression with shared parameters"?

I have paired values for three variables $z$, $y$, and $t$, and I wish the perform the regressions $z = g(t)$ and $y = f(t)$. I happen to know there is a bias in the variable $t$, thus the true times ...
Galen's user avatar
  • 9,421
5 votes
1 answer
200 views

Is the effectiveness of seemingly unrelated regression an example of Stein's paradox?

The existence of Stein's Example prima facie appears similar to seemingly unrelated regression (SUR) insofar as simultaneously estimating multiple parameters seems more effective than training them ...
Galen's user avatar
  • 9,421
2 votes
0 answers
280 views

Forbidden Regression Problem?

I am quite confused about the forbidden regression problem, and unsure if I have one in this case. Stage 1: Probit Model: Prob(Distress) = Set of regressors that predict vulnerability. The predicted ...
Kevin Smith's user avatar
1 vote
0 answers
42 views

SURE vs SEM: which one to use?

in my understanding, SEM should be applied in a case such as: \begin{equation} Y= \alpha_1 + \beta_1 X + \epsilon \end{equation} \begin{equation} X= \alpha_2 + \beta_2 Y + \epsilon \end{equation} as $...
Andrea's user avatar
  • 437
1 vote
0 answers
45 views

Assessing performance of entire nonlinear SUR in R?

I understand the McElroy's R-squared is used to assess the entire SUR system in Hamann & Henningsen's systemfit package in R. However, I've been running SURs ...
250gallontank's user avatar
1 vote
0 answers
112 views

How would you run a SUR with nonlinear constraints?

Chiappari et al (2012) estimate two Seemingly Unrelated Regressions like this: $$ y_1 = a_1 + b_{11}x_1 + b_{12}x_2 $$ $$ y_2 = a_2 + b_{21}x_1 + b_{22}x_2 $$ Next, they want to test the hypothesis ...
dash2's user avatar
  • 246
0 votes
1 answer
239 views

Deriving marginal effects for a bivariate model: ordered probit + linear regression

I am having trouble obtaining marginal effects of the following model in Stata, so that I would love to have some help in how to obtain an expression by hand: I have a system with two equations: an ...
Arthur Carvalho Brito's user avatar
0 votes
0 answers
58 views

Multivariate Multiple Regressions

I have 24 multiple regression equations (24 dependent series and 24 independent series) and want the coefficients to be equal. How do I combine them? I allow for correlation of errors and tried SUR, ...
Begum Sener's user avatar
0 votes
0 answers
79 views

T-test across two regressions (seemingly unrelated)

I want to compare across two regressions if the coefficients are equivalent using a t-test, how could I approach this and is there a function in R, which does this for seemingly unrelated regressions? ...
Question Anxiety's user avatar
0 votes
2 answers
838 views

Questions about Seemingly Unrelated Regression when all covariates are the same, but Y is different

Assuming I was estimating the same regression model, but had two dependent variables. Say one is income for women, and one is income for men. I want to understand if I understand the reason for using ...
Steve's user avatar
  • 681
0 votes
0 answers
42 views

VAR($p$) estimation

I am bit stumped by this result. Source: Remark on page 46 of Multivariate Time Series Analysis by Rsay. ... one can obtain the GLS estimate of a VAR($p$) model equation by equation. That is, one can ...
manav's user avatar
  • 329
2 votes
0 answers
179 views

What is the difference between SUR and OLS?

How is the output different in the SUR (Seemingly Unrelated Regressions) equation from the OLS? Is it still the same as Bo + B1x = y?
Hazel's user avatar
  • 21
4 votes
1 answer
444 views

Testing if coefficients are statistically significantly different across models

I will be building two zero-inflated negative binomial (ZINB) regression models, where each model is aiming to predict different disease count outcomes based on the exact same independent variables ...
geoscience123's user avatar
0 votes
0 answers
121 views

Is there any benefit of using GLS when the regressors are identical

I am reading Greene, Econometric Analysis, 7th Addition, I am seeking a point of clarrification. "The case of identical regressors is quite common [think a VAR mode].... In this special case, ...
Jack Salah's user avatar
1 vote
1 answer
2k views

Find (or calculate) log-likelihood value, AIC, and BIC for SUR model (for each equation) with systemfit

I have estimated SUR model with systemfit (R package). With the estimated results, I am trying to get logLik, AIC and BIC for ...
John legend2's user avatar
4 votes
1 answer
206 views

SUR and interaction terms

Suppose I want to determine if a simultaneous model (A) was identified: $y_1 = \beta_{10} + \beta_{11} x_1 + \beta_{12} y_2 + \epsilon_1$ $y_2 = \beta_{20} + \beta_{21} y_1 + \beta_{22} x_2 + \...
RegressForward's user avatar
0 votes
1 answer
519 views

Comparing OLS regression coefficients with SUR in SAS [closed]

I have the following code in Stata that I'm looking for the equivalent in SAS. Stata code: ...
user avatar
0 votes
0 answers
165 views

Simultaneous estimation of a group of linear model (regression) parameters

Suppose $y=ax+z$ where $x, y, z$ are random variables with range in $\mathbf R$, $\mathbf E[x]=\mathbf E[z|x]=0$ and $a$ is a constant. Note the distribution of $z$ conditioned on $x$ depends on $x$. ...
Hans's user avatar
  • 1,015
5 votes
1 answer
710 views

What is the difference between seemingly unrelated regression (SUR) and correcting a set of OLS results for multiple comparisons?

As I understand it, the the seemingly unrelated regressions (SUR) or seemingly unrelated regression equations (SURE) models estimate a set of Ordinary Least Squares (OLS) equations where the error ...
M. King's user avatar
  • 51
2 votes
0 answers
179 views

How to compare coefficients of multivariate multiple regression models, possibly using SUR

I am trying to compare the model parameters among three multivariate multiple regressions. All three models incorporate date from the same 97 individuals and share the same 4 independent variables (...
J. Joganic's user avatar
2 votes
0 answers
236 views

Solving SUR with (N-1) Equations

Problem I'm estimating a seemingly unrelated regression (SUR) with identical regressors where my dependent variable, $y_{cit}$, is a share of total across $c$, such that, $\sum_{c=1}^5 y_{cit}$ = 1. ...
Amstell's user avatar
  • 251
3 votes
1 answer
70 views

SEM One regressor depends on another regressor

I have the following structural model, by which one of the regressors is partially explained by another. $$ y_1= x_1+x_2+x_3+e \tag{1} $$ $$ x_1= x_2 + u \tag{2} $$ The questions are: a) Can ...
Luis's user avatar
  • 31
1 vote
0 answers
188 views

Are multivariate probit models with the same set of explanatory variables for each outcome more efficient that piecewise probit regressions?

I understand that multivariate probit models are analogous to SUR models. In the SUR case, there's no efficiency gain by fitting a SUR model over several independent OLS regressions when the model ...
David Pascall's user avatar
3 votes
0 answers
331 views

SUR with unbalanced panel and cross equation restrictions - available software?

I would like to estimate a SUR model with an unbalanced panel and with cross equation restrictions. This does not seem possible with the standard SUR commands in Stata or R. Do you have an idea which ...
clog14's user avatar
  • 231
1 vote
0 answers
29 views

testing hypothesis in spatial sure models

I have estimated spatial sur models (lag and error) using the spse program, spseml function. I want to test some restrictions about the coefficients in this model and I have not been able to do it ...
Pilar González's user avatar
0 votes
0 answers
150 views

CAPM Model - SUR Model -

We know that the CAPM regression (for a asset i) is given by $\ z_i = α_i1_T +$ $\beta_iz_m + ε_i = X∗θ_i + ε_i$ How can I show that the estimator of $\hat\alpha$ has the following distribution $\...
Bonsaibubble's user avatar
2 votes
2 answers
413 views

Why are individual fixed effects from a "within" and from a dummy variable panel model different?

I am working on a long panel data set with N=34 and T=132. I need to extract fixed effects from plm (within) model and estimation of the same model but including dummies for individuals (N-1) in OLS. ...
user138944's user avatar
1 vote
0 answers
386 views

Multiple impute multilevel data and postestimation tests

I want to do the following three things but am not sure how/whether they can be done: multiple impute multilevel data. I would be ok with just accounting for clusters, but I need to somehow account ...
Brett's user avatar
  • 141
1 vote
1 answer
101 views

Significant effect of control variable on the concerned determinant

I have a regression where I find the effect of x on y while controlling for z: y = x + z Theoretically, Both x and z have an effect on y. z has an effect on x. ...
Mumbo.Jumbo's user avatar
4 votes
2 answers
665 views

Books on Bayesian seemingly unrelated regression

I want to study seemingly unrelated regression using Gibbs sampling for many equations. Can someone suggest some books on Bayesian approach for seemingly unrelated regression (SUR) with R examples. I ...
Mumbo.Jumbo's user avatar
0 votes
1 answer
319 views

A model similar to vector autoregressive (VAR) model with different explanatory variables

VAR models do not allow the flexibility of having different explanatory variables in each equation. Are there any alternative models which allow this flexibility and written in a VAR form?
Thanos Sakkas's user avatar
1 vote
2 answers
121 views

Why the inverse of the covariance matrix is equal to the original covariance matrix in Seemingly Unrelated Regression?

In Zellner 1962 p350-351, you may see (2.4) & (2.6), and you can also verify that the Sigma(c)=Sigma(c)^-1. Why the inverse of the covariance matrix is equal to the original covariance matrix in ...
Veronique's user avatar
2 votes
0 answers
2k views

How to perform a SUR with Panel Data?

To do a sureg in Stata I was thinking of reg healthy x1 x2 ... i.year i.country ...
Maria Alegria Pinheiro Torres's user avatar
3 votes
1 answer
3k views

Difference between SUTSE (Seemingly Unrelated Time Series Equations) and SUR (Seemingly Unrelated Regressions)

I am studying time-series econometrics and in particular Dynamic Linear Models for multivariate time-series. Someone can help me in understanding which is the difference between SUTSE (Seemingly ...
PhDing's user avatar
  • 3,099
2 votes
1 answer
1k views

Test on SUR model

I have a SUR model with 22 equations, where each equation has the same 7 factors. I want to test if a coefficient (b3 in equation 1) is significantly different from another coefficient in another ...
a.paolini's user avatar