Formulae. Flashcards
Stratified sample =
(Strata / total) x sample size
Pearson capture - recapture:
( Num in 1st x num in 2nd ) / num in 2nd that are marked
Angles in a pie chart =
(360 / total frequency) x frequency for each class
Frequency density of histograms =
Frequency / class width
Mean of discrete data =
Sum of all values / number of values
Mean in a grouped frequency table =
Sum of (frequency x midpoint) / sum of frequencies
Weighted mean =
Sum of (weight x value) / sum of weights
Outliers =
Values > UQ + (1.5 x IQR)
OR
Values < LQ - (1.5 x IQR)
Skewness by inspection :
Positive skew - distribution is shifted leftwards
No skew - symmetrical distribution e.g. normal distribution
Negative skew - distribution is shifted rightwards
Seasonal variation =
Actual value - trend value
Mean point on scatter diagrams =
(Mean of x values, mean of y values)
Index number =
(Price / base year price) x 100
Weighted index number =
Sum of (index number x weight) / sum of weights
Chain base index numbers =
( Price / last year’s price ) x 100
Standard population =
( Number in age group / total population ) x 100
Standardised rate =
( Crude rate / 1000 ) x standard population
P(event) =
Number of successful outcomes / total number of outcomes
Expected frequency of event A =
P(A) x number of trials
Estimated probability =
Number of trials with successful outcomes / total number of trials
Risk =
Number of trials in which event happens / total number of trials
Relative risk =
Risk for those in the group / risk for those not in the group
Mutually exclusive events =
P(A or B) = P(A) + P(B)
Mutually exclusive, exhaustive events = (2)
P(A) + P(not A) = 1
P(not A) = 1 - P(A)
Probability of A or B in independent events =
P(A or B) = P(A) + P(B) - P(A and B)
Probability of A and B in independent events =
P(A and B) = P(A) x P(B)
Probability B given A conditional events =
P(B|A) = P(B and A) / P(A)
Three formulae used to check for independency in events =
P(A|B) = P(A)
P(B|A) = P(B)
P(A) x P(B) = P(A and B)
Standardised score =
(Score - mean) / standard deviation
Z scores =
(Value - sample mean) / standard deviation
Predicted value seasonal variation
Predicted value = trend line value + estimated mean seasonal variation
