Recent content by hlsmith

  1. hlsmith

    Hello

    Just take your best guess. We are not that picky.
  2. hlsmith

    AIC in ARIMA

    Can you examine residuals and their dispersion?
  3. hlsmith

    Does local linear regression include a weighting Kernel?

    Only kind of familiar with RDD, but wouldn't it be dependent on what you believe the underlying effect is (shapewise). I am also assuming you lose some interpretability with non-linear, but if it is appropriate it should probably be used. I may check out that paper tomorrow.
  4. hlsmith

    Significant treatment*pretest interaction, but interaction not shown with regression lines

    When examining an interaction term it is standard practice to keep the main effects terms in the model. What happens when you add them back in?
  5. hlsmith

    Hello

    Welcome to the forum. We look forward to your posts and contributions. What year are you in your program and what area of bioinformatics are you most interested in?
  6. hlsmith

    Difference between self-reported and actual measurements

    Well I was going to tell you to model the differences, but that seems to be out. You will need to look up what agreement stat you need for what I will loosely call ordinal data. It will be a derivative of the Fleischman Kappa inter-rater reliability stat.
  7. hlsmith

    More than one value in a single cell

    Did you pilot test your instrument before using it? Were the selections checkboxes our the respondent could type in text?
  8. hlsmith

    AIC in ARIMA

    Yeah that sounds right to me.
  9. hlsmith

    Beta Regression?

    Correct, for "count of a certain type of cell divided by the total number of cells" can be modelled with a count model. Do you have zero-inflated data? There is also the negative binomial model.
  10. hlsmith

    Relative standard error (RSE)

    Not sure of better ways. Could you just use a bootstrap. So get your above point difference estimate, then bootstrap the sample say a 1000 times and repeat the process for the point estimate. Or I don't know why you can just divide the 95% CI values by 797.
  11. hlsmith

    Is it possible to statistically analyse effect of independent variable on one of two interacting dependent variables?

    Could there be a mediated effect. Meaning phosphates direction effect colonization and indirectly effect colonization through biogrowth? Was treatment randomized?
  12. hlsmith

    Beta Regression?

    Likely, beta regression is a good fit. You can also plot the dependent variable using a histogram to visualize how close it is to '0'. If I recall, the interpretation of coefficients from beta reg is different from linear reg, so keep that in mind if your use it. What is your sample size and...
  13. hlsmith

    Beta Regression?

    Beta regression is fine for this. If the mean value and the majority of your percent DV is between 0.2-0.8, still running linear regression can be OK in some scenarios. The issues arise when values and precession estimates are close to the bounds (0 or 1).
  14. hlsmith

    Difference between self-reported and actual measurements

    Post a histogram of reported and actual and difference between then. So three histograms. Also post your sample size. Lastly, do you hypothesize the differences may vary based on any other variables?
  15. hlsmith

    Proability of failure

    Glad to hear you had 'success'! This will generate predicted probabilities, which can be used to interpolate values of your continuous variable within the range included the sample.