Please enable JavaScript.

Coggle requires JavaScript to display documents.

STAT1600 (7 time-series (7.1 time series data (time series data, time…

- - - - multiplicative time series model
      - moving average, centered moving average
      - ratio-to-moving average
        
        seasonal indexes(interpretation)
        
        seasonal adjustment (normalization)
      - deseasonalization
      - multiply unadjusted forecast values
        by appropriate seasonal index
  - - - appropriate when
      - smoothing coefficient
        
        interpretation
      - damping factor
- - - - notation & formula
    - - categorical: nomical, ordinal
      - continuous: interval,ratio
  - - - J-shaped, rectangular
      - positively/negatively skewed
      - bimodal, bell-shaped
    - - frequency table
      - pie chart & bar chart
      - box plot & side-by-side box plot
      - stem-and-leaf plot
      - dot plot
      - histogram
- - - - probability sampling
        
        simple random sampling
        
        stratified random sampling
        
        systematic sampling
        
        cluster sampling
        
        multi-stage sampling, multi-phase sampling
      - non-probability sampling
        
        convenience/haphazard sampling
        
        quota sampling
        
        snowball/network sampling
        
        respondent-initiated telephone polling
        
        internet survey
      - accuracy
        
        sample size
        
        bias
        
        selection bias
        
        non-participation(non-response) bias
        
        response bias
        
        margin of error(for 95% confidence)
    - - composite question
      - negative wording(do you agree that sb shouldn't--->do you agree or disagree that sb should)
      - offensive wording
      - sensitive question
      - ambiguous question
      - open questions with too few choices
      - answers not mutually exclusive
      - non-standard wording(slang)
      - loaded question with unjustified assumption(force the respondent to agree with the assumption)
      - leading question(suggest the respondent to answer in a particular way)
  - - - find what is confounding variable
      - completely randomized
      - block randomized
- - - - correlation doen't prove causation
      - only for linear relationship
      - effect of variability, discontinuity
      - correlation for combined data
- - - - probability
        
        logic
        
        long run relative frequency
        
        assumption
        
        direct observation (repetition)
        
        personal belief
        
        probability must be between 0 and 1& be coherent (logical, consistent, following rules)
    - - Rule 1
        
        totality
      - Terminology & Notations
        
        experiment
        
        outcome
        
        sample space
        
        event
        
        complement event
        
        subset event
        
        Rule 2
        
        mutually exclusive events
        
        Rule 3
        
        probability
    - - dependent & independent
      - conditional probability
    - - long-term gains & losses
      - expected value(EV)
        
        a long run concept
        
        ignores risk
        
        as mean number
  - - - bese rate fallacy
      - bayes' rule
        
        base rate
        
        sensitivity
        
        specificity
  - - - p-value
      - randomization test
      - permutation test
- - - - 95% confident that the probability is between... and ...
    - - SEM
      - difference in means
  - - - conditions & rules
- - - - percent, proportion, probability; risk
      - baseline risk, absulte risk, relative risk, increased risk
      - odds, odds ratio
      - rare events: relative risk (close to) odds ratio
    - - question of interest
      - row/column variable
      - conditional(row) percentage
      - rate per 100/1000/10000
      - row/column total
    - - population at risk
      - time period
      - need baseline risk
  - - - strength of relationship vs size of study
      - statistical significance & practical significance
    - - null & alternative hypothesis
        
        data snooping
      - compute the chi-square statistic
        
        expected cell count
        
        chi-square statistic
      - make the decision
        
        statistically significant
        
        confidence interval
      - p-value
    - - 1 determine the null hypothesis & the alternative hypothesis
      - 2 test statistic
      - 3 determine how unlikely the test statistic would be if the null hypothesis were true
      - 4 make a decision