Basics_1.txt

Q-Q Plot:
 quantile: lines divide data into equally sized group. ex, 0.75 quantile line- 3 quarters of the points are less than it.
step1: give each point its own quantile
step2: normal curve or any other distribution.
step3: add the same no. of quantiles to the curve as you created for the data. For the normal curve, "equal sized groups" means that there is equal probability 
       of observing a value within each group.
step4: plot q-q graph. Data quantiles vs normal quantiles.

Importance check for common distribution: assumption of a common distribution is justified or not when there are 2 data samples. If two samples do differ, it 
is also useful to gain some understanding of the differences. 
--------------------------------------------------------------------------------

Correlation coefficient: used to measure how strong a relationship is between data.
One of the most commonly used formulas in stats is Pearson's correlation coefficient formula. Two other formulas that are commonly used are sample and population
 correlation coefficient.
The pearson is not able to tell the difference between dependent and independent variables. 
--------------------------------------------------------------------------------

Outlier: It can be seen as a data point that is vey different from other observations. It can be caused while measuring. They can reduce the accuracy of a model 
therfore it's important to remove them from the dataset.
Z-score can be used to detect outliers. But for this data must be normally distributed. 
IQR: IQR is difference between 3rd and 1st quartile. Identify if a point is an outlier if it is less than q1-1.5*IQR or greater than q3+1.5*IQR. 
Other methods include DBScan clustering, Isolation Forests, and Robust Random Cut Forests.
--------------------------------------------------------------------------------

Central Limit theorem: It states that the sampling distribution of the sample mean approaches a normal distribution as the sample size gets large no matter what 
the shape of the population distribution. Used in hypothesis testing and calculating confidence intervals.
--------------------------------------------------------------------------------