8 - Normal Distribution, Probability And Statistical Significance Flashcards
Negatively skewed
Cloche à droite
Positively skewed
Cloche à gauche
Characteristics of normal distribution
- the distribution is bilaterally symmetrical and bell-shaped
- mean, median and mode coincide
- tails never reach x-axis
- range differs because STD differs
- Kurtosis = degree of peakedness
- skewness = 0 ; peakedness = 3
Negative z score
Indicate that the value is below the mean
Z score
Provide a standard unit of relative worth that enable us to compare across different distributions.
Slew type transformation
Moderate: sqrd
Substantial: ln
Extreme 1/x
Probability of value between STD
68,26% - 1SD
95% - 1,96SD
99% - 2,58SD
99,9% - 3SD
Testing weather a distribution is normal methods
- calculate skewness and kurtosis
- QQ plot
- Kolmogotov-Smirnov and Shapiro-Will tests
Probability
Numerical measure of the likeliness of an event.
In practice event are neither impossible nor absolutely certain
Level of significance
Or alpha level is the o value that defines very unlikely events. Alpha is the probability of Type I error