Please enable JavaScript.

Coggle requires JavaScript to display documents.

Quantitative methods - Coggle Diagram

- - - - Non-numerical/Numerical
        
        Nominal
        
        counting frequency
        
        Ordinal
        
        eg. rate hotel star
    - - Numerical
        
        Interval
        
        doesn't have to be 0 : eg.temperature (artificial 0)
        
        Ratio
        
        has absolute 0 : eg. money, time (no money in the bank)
- - - - Mean:avg
      - Median: use when have extreme values
      - Mode
      - Percentiles : how data spread over interval i = (p/100)/n
        
        i not integer, round up : pth =value in ith position
        
        i = integer pth = avg of positions i and i+1
      - Quartiles: 2nd quartile = 50th percentile = median, third=75th percentile
    - - Range : largest-smallest data
        
        Interquartile Range :dif between third & first quartile
      - variance : avg of (each data value-mean)^2, the smaller, the less risk
        
        standard deviation : s= root s^2
        
        coefficient of variation : how large sd is to mean , (s/mean of x *100)%
    - - skewness
        
        symmetrical : mean = median, skewness=0
        
        Moderatly skewed left : mean<median, skewness=negative
        
        Moderately skewed right : skewnes=posittive, mean>median
      - z-scores : standardized value (z=(xi-xbar)/s)
      - Empirical Rule : Bell shaped (normal distribution)
      - Outliers : z score<-3 or >3
      - Chebyshev's theorem (1-1/z^2) will be within z score
    - - weighted mean
- - - - Complement of an Event : me out of circle
      - Union of two events : A u B
      - Intersection of two events: A intersect B
      - Mutually Exclusive events : no common sample points [boobs]
      - Conditional probability : prob(A|B)=(prop A intersect B)/prob B
      - independent events: p(A intersect B)=p(A)p(B) [rolling dice first and second time = independent]
      - Multiplication law : prop (A intersec B) = prob B * conditional A|B
      - Bayes' Theorem