Search results

  1. hlsmith

    design question

    You model propensity scores and put those into the outcome model as weights. The outcome model is typically a simple regression, exposure and outcome. Can you provide a histogram of the outcome variable.
  2. hlsmith

    Immigrants VS Americans educational levels in relation to Welfare

    Yeah I thought you were likely human or a tricky bot. Good to hear you aren't metal/silicon!
  3. hlsmith

    design question

    If I was in your shoes, it seems like of those eligible, you could run a propensity score model to get weights to use in your outcomes model. As you mentioned, you may not have access to all the variables but this would be better than a naive or conditional model. For the propensity score model...
  4. hlsmith

    Scree plot interpretation

    I have never seen this, but I wonder of internal validation could be use via having a holdout set to confirm cut-offs. I haven't reviewed many of these, but yes 2 seems to be the blatant and magical number in this one.
  5. hlsmith

    Chi Square question

    Please describe in more detail what these data represent. I would calculate the proportion of with the event and calculate a confidence interval around it (binomial data). But without more information I am just guessing at the context and objective. Say we were talking about males vs females...
  6. hlsmith

    Regression Analysis

    Time series or random effects may be of interest, since observations across years are correlated. For example, my weight today is correlated with my weight last week and to a lesser extent the week prior. Ignoring this dependency structure across time can lead to biases in the interpretation of...
  7. hlsmith

    Immigrants VS Americans educational levels in relation to Welfare

    Hey @Dick_ mob1 - what is three minnus 2? Just checking you aren't a bot given your posting on old threads and having such a cool avatar picture.
  8. hlsmith

    Appropriate Sample Size

    Alright, for a sample size calculation you should pick the closes (smallest difference)comparison you can think of that you are interested in and power for that. All other tests should then be reasonably powered. You can validated the results using the data you didn't sample from. HOWEVER, if...
  9. hlsmith

    Appropriate Sample Size

    Yeah if you have the population use it. Run analyses on a server if at an university, if you are using a local machine run it in parallel. If you don't have multiple cores run it over night. Not using the full data is running from the real problem, your inability to truly answer your questions...
  10. hlsmith

    Appropriate Sample Size

    You word this as though you have access to the full population data. If so, why not use the who thing? In both scenarios, you just use random sampling without replacement. Boom you are done. Is there a particular question you are trying to power using these data or a certain level of...
  11. hlsmith

    Survival analysis with R Commander

    What are you trying to explicitly model? Please provide an example of what your data may look like. In a basic survival model you have an exposure (binary treatment: 0,1), censoring variable (person censored: 0,1), and time (typically a continuous variable until person had event or was...
  12. hlsmith

    What would be two/three good variables to look at for creating a linear regression model?

    If this is an assignment, then yeah, an r = 0.8 is fairly high in most contexts.
  13. hlsmith

    unused arguments

    Sometimes if cutting and pasting was used, your quotation marks may be backwards or not registered correctly. You can also try rewriting the line of code from scratch and seeing if that works.
  14. hlsmith

    Help with linear and unconstrained MMSE estimators

    a) looks like it is just requesting you model a least squares (simple linear regression). As for b.), I am not familiar enough to know what they are looking for, polynomials?
  15. hlsmith

    Just for laughs!

    What does the 'B' in Benoit B. Mandelbrot stand for?
  16. hlsmith

    Hierarchical regression building support

    By block do you mean level? For clarity we are talking about data clustered in units, right?
  17. hlsmith

    Probability puzzle

    I kind of want to change it to a dice rolling problem since my mind won't let me imagine race conditions can be exactly replicated. However - this problem doesn't discuss ties. Hmmm
  18. hlsmith

    Baffling data entry error

    So it was not creating anything for mydata? Did you get an error?
  19. hlsmith

    help with pre /post design and different subgroups

    Why do you need a cutoff value? Categorizing continuous variables just results in a loss of information and doesn't allow you to examine non-linear effects.
  20. hlsmith

    Multiple comparisons vs multiple pairwise t tests

    I feel like qvalues may be used in these settings.