Regression

Analysis
Regression Analysis
- Statistical tool
- Estimates the relationships among variables
- Assembled data
- Economic Statistics (econometrics)
- Prediction and Forecasting
- Dependent and Independent Variables
Methods
1. Parametric Regression
- Finite
- Not flexible
2. Non-Parametric Regression
- Infinite
- More flexible

Commonly used approaches:

1.Smoothing Splines
2.Kernel Regression
Least Square Formula

Other Methods
1. Bayesian Method/Bayesian linear regression
an approach to linear regression in which the
statistical analysis is undertaken within the
context of Bayesian inference.
2. Percentage Regression
Example: The number of officers on duty in a Boston city park and the number
of muggings for that day are:
Officers Muggings
10 5 calculate the regression line for this data,
15 2 and the residual for the first observation,
16 1 (10,5). What percentage of variation is
1 9
explained by the regression line?
4 7
6 8
18 1
12 5
14 3
7 6
Solution:

We found that

m= = - 0.493 and

b = - (-0.493) * = 9.780
Then, the equation of the regression line is
= - 0.493x + 9.780
3. Least absolute deviation
Also known as Least Absolute Errors,Least Absolute Value is a
statistical optimality criterion and the statistical optimization
technique that relies on it. Similar to the popular least squares
technique, it attempts to find a function which closely
approximates a set of data.

4. Non- parametric regression

Is a form of regression analysis in which the predictor does not
take a predetermined form but is constructed according to
information derived from the data. It requires larger sample
sizes than regression based on parametric models because the
data must supply the model structure as well as the model
estimates.
5. Distance metric learning
the task of learning a distance function over
objects. A metric or distance function has to
obey four axioms: non-negativity, identity of
indiscernibles, symmetry, and subadditivity/
triangle inequality.