Search results

  1. staassis

    Time Series books

    I think this is an interesting topic. I did put my 2 cents in but would anybody else have anything to contribute?
  2. staassis

    Probability of regression result

    If it's not an idealized hw problem, then linear regression is far from being appropriate. First, you need to transform the data into returns, log-returns or some other weakly stationary stochastic process. Second, you need to apply time series techniques. In particular, as ondansetron pointed...
  3. staassis

    To assess meaningfulness between an independent variable and a dependent one, is just the p value alone adequate and/or the correlation?

    The two characterizations are separate: 1] "The effect is statistically significant" means that the true parameter has a value different from the value in the null hypothesis. In the modeling setting, this typically implies that the corresponding predictor has a relationship with the dependent...
  4. staassis

    Which statistical test to use

    "I feel it is sufficient to answer the question..." You feel? I thought your coming to this forum was a demonstration of understanding that you do not know statistics sufficiently well...
  5. staassis

    Probability of regression result

    This is a hw problem. You can show some effort by posting your thoughts or you can just google for "linear regression prediction interval".
  6. staassis

    Mixed Modeling for Longitudinal Data

    To identify the "optimal model" in a specific class of models, you can use one of the standard model selection criteria. For example, you can 1] make sure that all the parameters are statistically significant, 2] the "optimal model" has the lowest Akaike Information Criterion score (AIC) or...
  7. staassis

    Dont know which statistical test I should use?

    Or λ(t) = exp( B0 + B1 * t + B2 * t^2 + B3 * Glucose(t) + B4 * log(Glucose(t)) + ... ).
  8. staassis

    Dont know which statistical test I should use?

    Simplest methods of survival analysis, like Kaplan-Meier estimator or Cox model, would not be applicable here. The reason is we have a continuous predictor (glucose concentration) which changes randomly over time. The survival of each cell is best characterized by a counting process with...
  9. staassis

    Measuring dependence over time?

    You can transform the variables into stationary versions and then estimate a Vector Autoregressive Model.
  10. staassis

    Which statistical test to use

    Too little data.
  11. staassis

    Time Series books

    The following book contains concise and efficient summary of VAR but nothing too detailed. In general, the book is a useful complement to a typical time series book since it discusses regimes-switching models, Bayesian methods for stochastic processes, etc. Lai & Xing (2008). Statistical Models...
  12. staassis

    time series analysis

    No reason at all for the optimal window size to be the same. Example: A_i(t) = M_i(t) + E_i(t), i = 1,2 where M_i(t) is reset to a random value every S_i days, E_i(t) is white noise, independent of everything else. Now, imagine that S_1 = 10, S_2 = 20. Then the...
  13. staassis

    time series analysis

    Again, "it is not clear how you determined the window size". Also, you were saying: "I couldn't logically think a way...". Nobody is asking for an abstract theoretical argument here. The optimal model for any stochastic process (A, B or C) must be dictated by data.
  14. staassis

    time series analysis

    It is not clear how you determined the window size, among other things. Also, what's optimal for the aggregate measure does not have to be optimal for the constituents. The dynamics of an aggregate measure is typically better behaved than that of its constituents.
  15. staassis

    Standardizing Performance (sports-related)

    Yes, and to know how to adjust for the quality of opponent you need a model.
  16. staassis

    Standardizing Performance (sports-related)

    The suitable methods depend on how much data you've got. At the very least, you can look into 1) logistic regression for the probability of victory in each game and 2) Poisson regression and negative binomial regression for the points scored in each game. However, much more is possible in...
  17. staassis

    how to determine whether the point is outside/within the ellipse in PCA

    Sorry, I do not understand what the problem is. Just calculate the Mahalonobis distance and if it's above the 80-th percentile paint the point into the red color... But hey, you've done that already.
  18. staassis

    No idea what statistical tool to be used. Need help

    None. Too little data.
  19. staassis

    Growth Rate/trend identification

    The model would depend on how much data you have.
  20. staassis

    Time Series Analysis

    So send your draft, show the effort and we'll tell you where you are wrong.