Search results

  1. staassis

    Need help evaluating a PCA

    You do not. Factor loadings (if using the traditional definition) tell you how to represent the original variables in terms of factors. They do not tell you the reverse: how to calculate factors in terms of the original variables. The easiest approach is saving the factor scores (in SPSS, R...
  2. staassis


    Yes, you can. Choose the optimal penalty coefficient (λ) using leave-one-out cross-validation. It is likely to be substantial.
  3. staassis

    What statistical test for my data?

    You have to build a generalized linear model (GLM) of the form: Var3 ~ Var1 + Var2 GLM types to consider: Poisson regression, negative binomial regression, Poisson regression + zero-inflated component, negative binomial regression + zero-inflated component. You can choose the "optimal" GLM...
  4. staassis

    Is "time analysis" possible here?

    Yes, you can study the time effects using a panel-data model. The following framework must work in your case: DV_ij = α_i + β_1 * Time_j + β_2 * Time_j^2 + γ_1 * X_i1 + ... + γ_p * X_ip + ε, where DV_ij is the dependent variable for participant i and survey time j. You can consider two...
  5. staassis

    Historical research question

    Which photos? Nothing got attached... Aside from that, extensive attachments are bad style. You are asking for an advice, not for somebody to look deeply into your work and perform formal consulting... Also, please make your textual description shorter, straight to the point. Thank you.
  6. staassis

    What package to install

    Fair enough. Pretty great people go a long way towards something great in life.
  7. staassis

    Covid-19& presidential election analytics

    I am very sad, time after time seeing people who think they came up with an original and topical data analysis question: Covid-19.... There is too little data. Statistics is the science about what to do with data. One needs data.
  8. staassis

    What package to install

    @Dason, why is Iowa best? Any advantages over CA-1 or PA-1? Very curious.
  9. staassis

    What package to install

    The PA-1 mirror has always worked well for me. I believe it's Pennsylvania.
  10. staassis

    Clustering of variables with time and grouped data

    If we have repeated measures, it is still ok to apply clustering to variables. It would not be ok to apply clustering to observations. Variables are like people, going through all kinds of related and unrelated situations. And some people remain friends through life, and some people don't.
  11. staassis

    Regression being a sum of two regressions

    Such problems are typically addressed by building one time series model using, say, hourly data. Such model would contain daily seasonality, capturing different demand during different times of day. What you have done seems to ignore the early morning and early evening dynamics, providing...
  12. staassis

    What statistical test for my data?

    What are your research questions?
  13. staassis

    Historical research question

    Please expand.
  14. staassis

    Clustering of variables with time and grouped data

    I'm afraid, I do not see what the problem is. Testing subjects in different environments ensures variability in the data, which allows us to see even better which variables belong together and which variables do not. So the accuracy of cluster analysis is increased precisely because we have...
  15. staassis

    Exponential Distribution

    Part a) is correct. For part b, use the fact that Z is the interarrival time of the Poisson process which is the sum of the Poisson process corresponding to X and Poisson process corresponding to Y.
  16. staassis

    Two categorical and one continuous dependent variables + one categorical independent variable. Which test should I run?

    Still do not understand variable "No. of followers". What are your reasons for considering the variable categorical?.. Is the variable defined for each post. My understanding is that, in your analysis, post is the observation unit, i.e. 1 observation = 1 post.
  17. staassis

    Which test of association can I use?

    The default choice is Mann-Whitney test. If the sample size is substantial in each of the two groups ("yes" and "no"), you can also use the independent samples t-test. The two tests are likely to agree. If the relationship between the two continuous variables is linear you can use the t-test...
  18. staassis

    Is this matematical problem also related to some statistic predictive model problem?

    To the best of my knowledge, the question above cannot be rephrased as some well-known statistical problem. However, it can be solved with a statistical method known as Monte Carlo.
  19. staassis

    The effect of curriculum change on nutrition behavior, problem with regression function

    In a way, in a sense, "fictional" data analysis is a contradiction in terms. The correct data analysis begins with the questions: "Which data are available?" "Which variables are available? "Can I collect even more variables by administering a controlled experiment?"... Unfortunately, if we are...
  20. staassis

    Comparative test to use

    To see if there is a relationship between bicycle type and helmet usage, you can run chi-square test for independence... You can also follow @hlsmith's suggestion and run logistic regression of the form Helmet Usage ~ Bicycle Type A + Bicycle Type B, where 1) one of the bicycle types is left...