Term
|
Definition
| Q3-Q1 and contains the middle 50% of observations |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
| In a positive skewed distribution.. |
|
Definition
| You have skew towards lower values mean>median>mode |
|
|
Term
| In a negative skewed distribution.. |
|
Definition
You have skew towards higher values
mean<median<mode |
|
|
Term
| In a box-plot, if the median is closer to Q1, then we can say it has a__________skew |
|
Definition
|
|
Term
| In a box-plot, if the median is closer to Q3, then we can say it has a__________skew |
|
Definition
|
|
Term
| What are the basic characteristics of probabilities? |
|
Definition
1. Values 0-1
2. Sum of all probs. =1
3. Mutually exclusive
4. Independent events |
|
|
Term
| What is the addition rule? |
|
Definition
| P(A OR B)= P(A)+ P(B) - P(A and B) |
|
|
Term
| What is the multiplication rule? |
|
Definition
P(A AND B)= P(A|B) P(B) general
P (A) * P(B) if independent |
|
|
Term
| What is the difference between a combination and a permutation? |
|
Definition
| In permutation, the order of objects matters |
|
|
Term
| What is the binomial distribution? |
|
Definition
| Probability of outcomes for a series of random events each of which can have only two values |
|
|
Term
| What is the central limit theorem? |
|
Definition
| Given a pop. of any non-normal distribution, the sampling distribution of the sample mean, computed from all possible values of size n from this pop. will be approx. normal |
|
|
Term
|
Definition
| A statistical technique for comparing means for multiple populations by partitioning total variation |
|
|
Term
|
Definition
| A global test for the Ho of equality of all means |
|
|
Term
| What is the Bonferroni correction? |
|
Definition
| A method to deal with multiple comparisons. Modify significance level a*=a/(p 2). Where p= # comparisons and a the desired alpha level |
|
|
Term
| What is the coefficient of determination (r2)? |
|
Definition
| The proportion in the variability of Y explained by X |
|
|
Term
| How should residuals look on an histogram if assumptions hold? |
|
Definition
|
|
Term
| How should a plot of residuals vs. X look if assumptions hold? |
|
Definition
|
|
Term
| What are the assumptions in linear regression models? |
|
Definition
1. Linear relationship
2. Independent distribution
3. Normally distributed
4. Error terms ind. and normally distributed |
|
|
Term
| What is the AIC used for? |
|
Definition
| To compare several MLR models with possibly differing numbers of predictors (not nested). Choose model with lower value |
|
|
Term
| How do you detect possible confounding in a regression? |
|
Definition
| When b1 is diff. (>10-15%) in a SLR and a MLR and confounding criteria are met |
|
|
Term
| How do you detect possible mediation in a regression? |
|
Definition
| When b1 is different in a SLR and MLR and is in the causal pathway |
|
|
Term
| How do you detect possible effect modification in a regression? |
|
Definition
| When the Y-X relation varies by the level of C. Then C modifies the effect of X on Y |
|
|
