Techniques for analyzing the relationship between one (or more) "dependent" variables and "independent" variables.
0
votes
0answers
9 views
Goodness of fit vs. significance
If you had two models - one with a better fit, and one with predictors of higher significance - which one would you choose?
Keep in mind that both models are considered a good fit. One has a 3% ...
1
vote
0answers
11 views
Nested Logit model in r
I am trying to run a nested logit using mlogit in R to analyze data from choices people made. There are 4 possible alternatives they could choose from, but in any given choice situation a person had ...
0
votes
0answers
20 views
Semantic Dominance from similarity measures
I was reading a research paper on image annotation.
The paper gives an example in which similarity measures between a set of words are given and then semantic dominance of each word is calculated.
...
1
vote
0answers
49 views
Regression coefficient equivalence
In the exercises 5.1 from Meyers "Classical and modern regression with application" ask you to prove that:
1- The regression coefficient associate with $X_{j}$ in the multiple regression of $y$ ...
-1
votes
0answers
27 views
Regression Analysis Question
Assuming Congo gets foreign aid from Denmark, Sweden, UK, US and Germany. How do I go about getting the effect of Denmark's aid on Congo. I am currently running a fe panel regression, but I am afraid ...
0
votes
1answer
38 views
How should I interpret the result of my OLS (involving H-statistic)?
Please consider the following excerpt from "Measuring and Explaining Competition in the Financial Sector" by Bikker and Spierdijk (2008):
"In many economic theories, competition is related to the ...
2
votes
1answer
77 views
Diagonal elements of the projection matrix
I am having some problem trying to prove that the diagonal elements of the hat matrix $h_{ii}$ are between $1/n$ and $1$.
Suppose that $Range(X_{n,k})=K $ the number of columns of our matrix of data ...
1
vote
2answers
56 views
Regression and point-specific p-values (using R for explanation)
Consider the following (in R):
library(MASS)
plot(stack.loss~Air.Flow,data=stackloss)
regression <- rlm(stack.loss~Air.Flow,data=stackloss)
abline(regression)
...
0
votes
0answers
27 views
Percentage change vs. first difference
I want to regress (using an OLS regression) survey results on stock returns. How do I decide whether I use the first difference or the percentage change of the survey results?
0
votes
1answer
61 views
Covariate vs. factors
I have a survey problem where the dependent variable (ordinal) is in Likert-type scale (i.e. 1 to 5 from most satisfied to most dissatisfied) and two sets of independent variables. One set has 7 IVs ...
0
votes
1answer
38 views
How to understand SARIMAX intuitively?
I'm trying to understand a paper about electric load forecasting but I'm struggling with the concepts inside, specially the SARIMAX model. This model is used to the predict the load and uses so many ...
0
votes
0answers
17 views
Does doing predictive regression need a smaller sample size than exploring risk factors with logistic regression?
Cardiologists have a tool called EUROScore used to adjust the risk associated with performing heart surgery. It comes into play when, for example, one surgeon is recognised as being more expert and so ...
4
votes
1answer
35 views
Justifying the distribution for the maximum likelihood estimator in a linear regression example
Data $(x_1, y_1), \dots, (x_n, y_n)$ is modelled with $x_i$ being non random and $y_i$ being observed values of $$Y_i = \alpha + \beta (x_i - \bar x) + \sigma \epsilon_i$$ with $\epsilon_i \sim ...
2
votes
1answer
28 views
Testing a relationship between a continuous predictor and binomial outcome
I am testing a relationship between a continuous predictor and a binomial 0-1 outcome variable. My hypothesis is that the smaller the value of the predictor the more likely it is to fall into group 0 ...
0
votes
0answers
26 views
PROC CORR vs. REG?
Do both my independent and independent variable have to be normally distributed to run a linear regression? My dependent variable is normally distributed, but my independent is not...therefore, ...
2
votes
2answers
52 views
Which is applicable, ordinal or multinomial regression model?
I have done a job satisfaction survey where the DV is a 7 point Likert type scale and 5 IVs with 6 point Likert-type scale and 6 IVs with 5 point Likert type scale, all ordinal.
Which type of ...
3
votes
0answers
44 views
Generalised least squares: from regression coefficients to correlation coefficients?
For least squares with one predictor:
$y = \beta x + \epsilon$
If $x$ and $y$ are standardised prior to fitting (i.e. $\sim N(0,1)$), then:
$\beta$ is the same as the Pearson correlation ...
9
votes
1answer
130 views
Does stepwise regression provide a biased estimate of population r-square?
In psychology and other fields a form of stepwise regression is often employed that involves the following:
Look at remaining predictors (there are none in the model at first) and identify the ...
1
vote
0answers
45 views
Why does k-NN perform better than SVR and linear regression?
I have a data set used in a regression with 30 attributes and 30K instances. I am trying out a bunch of algorithms (SMO regression, Linear Regression and K-NN) but it was quite surprising to see that ...
1
vote
1answer
36 views
Interpretation of interaction effect in negative binomial regression
I am trying to interpret my interaction effects, which are all negative.
One example:
Experience (variable A) x absolute size of the acquired knowledge base (variable B): B= -0.002, exp(B)= 0.998.
...
1
vote
0answers
18 views
Discretizing all attributes vs. discretizing only the class label
I have a numeric dataset and I would like to apply classification algorithms on it. So I am using Weka's Discretize filter to convert the numeric values to discrete. However I wonder what can be the ...
0
votes
0answers
15 views
Panel data by subgroups
Panel data by subgroups
I have a panel dataset on 200 firms showing the turnover of financial managers and firm performance for ten years.
Some of the individuals are fired and some get promotion ...
3
votes
2answers
83 views
Linear vs. nonlinear regression
I have a set of values $x$ and $y$ which are theoretically related exponentially:
$y = ax^b$
One way to obtain the coefficients is by applying natural logarithms in both sides and fitting a linear ...
0
votes
0answers
23 views
Transforming t-distributed residuals to normal
I have an AR process with t-distributed noise at each step. Does anyone know a transformation i can apply to get approximately normally distributed residuals?
Thanks!
6
votes
3answers
175 views
Linear regression with slope constraint
I want to perform a very simple linear regression in R. The formula is as simple as $y = ax + b$. However I would like the slope ($a$) to be inside an interval, ...
2
votes
1answer
28 views
What value is conditional logistic regression if two cohorts are already matched on everything of interest?
My situation is this:
I am comparing two cohorts
I have matched the two cohorts on all of the factors that I am interested in
The two cohorts are already balanced on all of the factors that I would ...
1
vote
0answers
16 views
Use GLM weights to regularize noise
I do a GLM containing 8 predictors on a multivariate data set. Six of these predictors encode effects that have actually been manipulated in my experiment (effects of interest), the other two ...
2
votes
0answers
52 views
Regression model where errors are not normally distributed
From a Physics equation I have the following model:
$$W=\beta_0+\beta_1Z$$
$\beta_0$, $\beta_1$ are fixed values for which I want to find a $1-\alpha$ confidence region.
I have ...
1
vote
2answers
47 views
Understanding and interpreting consistency of OLS
Many econometrics textbooks (e.g. Wooldridge, "Econometric analysis...") simply write something similar to: "If the population model is $y = xB + u$ and (1) $\text{Cov}(X,U) = 0$; (2) $X'X$ is full ...
3
votes
1answer
34 views
Given the below dependent variable description, should I chose either Ordered or Multinomial logistic regressions?
My dependent variable looks like a range of ranks. It actually might be considered that way. But the ranking is based on subjective non-quantifiable cutoffs. We assessed the behaviors of a group of ...
0
votes
0answers
19 views
Looking for a regression model or other statistics models for competitions
We have botanical data and need to analyze this sort of scenario, in terms of students for the simplicity of explanation:
We have a series of competitions or trials, each one involves two individuals ...
0
votes
0answers
19 views
Heckman Sample Selection - exclusion restriction
Clearly there are some things I have not yet understood about sample selection and the application of the two-stage Heckman model. I hope someone can help me understand.
Say you have data on ...
1
vote
0answers
35 views
What kind of attribute selection methods exist for high-dimensional regression data?
I have a 10,000 dimensioned dataset where all attributes are numeric values. I would like to select the best e.g. 50 attributes out of 10,000 so that I can run regression algorithms on it.
I've tried ...
1
vote
3answers
66 views
Strange outcome when performing nonlinear least squares fit to a power law
I have a data set (given below in my MATLAB code) y vs. x and my eventual goal is to fit it to a power law $y=ax^b$ to see what exponent $b$ I get. I did some non-linear least squares fitting and ...
5
votes
1answer
78 views
How many observations per subject are necessary to fit a random slope in a mixed model?
I am working on a project that collected data retrospectively on subjects. There are subjects with multiple points of follow up per person, anywhere from 1 to 3 measurements. The timing of such ...
0
votes
0answers
20 views
how to change fused lasso regression depict image
We performed aCGH and depict the result using fused lasso regression. We got similar image as shown in below:
Since we have more samples, the left side of so called legend (Gain or loss) came for ...
1
vote
0answers
5 views
Quadratic terms in regression analysis [migrated]
Quadratic terms are quite common in regression. Here is an example from John Fox (http://www.jstatsoft.org/v08/i15/paper)
...
1
vote
0answers
39 views
linear regression terminology question
I begin my journey with the subject of regression. I have this basic question: I have values of factors of a model, together with values of residuals, and I need now to regress the factors against the ...
0
votes
1answer
58 views
How to fit (linear regression) a ratio of all independent variables?
I am trying to find the best fit between an species dataset and prevailing climatic conditions, in order to be able to predict the environmental conditions from the species dataset (paleoclimate ...
1
vote
3answers
85 views
Skewed data for regression analysis
I have 30 features in my self-collected dataset where I want to build a regression model. When I look at my data, most of the attributes (95% of the data points) are skewed on a very small range. Out ...
1
vote
1answer
57 views
Is it reasonable to compare a regression model with machine learing algorithms using RMSE?
I have a 70K x 30 dataset and I want to build a regression model on it. Right now, I am running a bunch of algorithms via Weka tool with cross-validation and I compare the RMSE values reported by Weka ...
2
votes
1answer
36 views
Nonsignificant interaction still causes main effect to flip?
This is my first post so forgive me if this is question has already been asked. I searched around the forum but didn't find anything specific to what I'm looking at currently.
I'll keep things as ...
0
votes
0answers
15 views
Standardized betas coefficients for second order models (quadratic)
I fit a quadratic model with one interaction term. In basic models, the standardized coefficient with absolute value closest to zero is the most important predictor.
However, I got two terms with ...
0
votes
0answers
21 views
Subset risk ratios are not collapsible
I have a monthly data that associates certain pollutants exposure zone (var:exp) to the monthly counts of ari (acute respiratory diseases)...
2
votes
0answers
27 views
Finding correlations between aggregated and non aggregated data
I am trying to correlate server usage data with CPU utilization. E.g., X number of transactions on the server correlates with Y amount of CPU utilization. The transactions are reported hourly and ...
1
vote
1answer
69 views
Interpretation in log linear regressions with coefficients bigger than 1
Just to clarify something regarding coefficients bigger than 1 in log-linear regressions.
If we have this regression, how would we go about to interpret the 1.12?
D1 is dummy variable for having a ...
11
votes
1answer
58 views
When building a regression model using separate modeling/validation sets, is it appropriate to “recirculate” the validation data?
Suppose I've got an 80/20 split between modeling/validation observations. I've fit a model to the modeling data set, and I'm comfortable with the error that I'm seeing on the validation data set. ...
3
votes
1answer
52 views
Error propagation with linear regression
I'm trying to obtain an estimation of the uncertainty related to an analytical method:
my function is just a linear regression $f: y=ax+b+ \epsilon$ with $y_i=\frac{R_i}{C}$, both $R$ and $C$ are ...
1
vote
0answers
32 views
Regression for foreign aid effectiveness
I'm trying to use a multiple regression (in Excel) to determine the effectiveness of foreign aid on GDP growth. I've forgotten everything since college...
My plan is to run a regression using ...
1
vote
1answer
44 views
Smart way to search through a very large parameter space
I have a system whose performance is based on a rather large parameter set (200 parameters, lets say, of which each can take a very wide range of values). There are tests to evaluate the performance ...
