linear model

  1. E

    Linear regression

    Hi, I have fit 100000 linear models, M ~ age + age^2, where M is a frequency. I found 5,000 models were significant, but would like to narrow it down to find the most ‘extreme’ cases. Is this possible? Does anyone have any suggestions or any references that may help?
  2. M

    Estimate the difference between intercepts and slopes among multiple cross-sectional data collected at different time points

    I am trying to do some analysis using 3 DHS dataset and check if there are differences between both intercept and slope at different survey rounds. What kind of modelling do I have to use? References to resources will also be highly helpful as I will be reading around this stuff. Thanks in advance!!
  3. C

    Linear mixed-effects model - How to compare absolute values of interaction slopes

    Hello I have built a linear mixed effect model in order to compare repeated measures (corresponding to 2 successive experimental conditions) in 2 different groups. According to this model, my outcome was significantly predicted by the group, the condition and the group x condition...
  4. C

    Maximum likelihood estimator in linear model

    Consider two simple linear models. y_{1j}=\alpha _1+\beta_{1}x_{1j}+\epsilon_{1j} [\math] and y_{2j}=\alpha _2+\beta_{2}x_{2j}+\epsilon_{2j} , j=1,2,...,n>2 where \epsilon_{ij}~N(0,\sigma^2) I have two questions. Obtain the maximum likelihood estimators of sigma^2...
  5. D

    R linear model on big table

    Hi all! I have a big table and want to run a linear regression (or other linear model) in R, but not sure how to do that.... this is a very short view of what I have (actually I have got about 5000 sample) Latitude sample48109_1 sample48326_1 sample48827_1 59.8 21 29 129 59.8 18 23 32...
  6. C

    R-squared distribution

    Suppose I have a simple situation where there is one quantitative explanatory variable and one quantitative dependent variable. If I construct a linear regression for each of a set of samples, what would characterize the distribution of the r-squares of my lm's?