DATA ANALYTICS

DATA ANALYTICS - SCIENCE

The Journey towards learning about Data Science, is interesting and mind-cracking. Where to start and what are the topics to be covered, and many more questions to be answered. Trying to demystify these details as much as possible. These will not be the exhaustive ones and there will be many more and different approaches.

Disclaimer:

• This document contains unedited notes and has not been formally proofread.
• The information provided in this document is intended to provide a basic understanding of certain technologies.
• Please exercise caution when visiting or downloading from websites mentioned in this document and verify the safety of the website and software.
• Some websites and software may be flagged as malware by antivirus programs.
• The document is not intended to be a comprehensive guide and should not be relied upon as the sole source of information.
• The document is not a substitute for professional advice or expert analysis and should not be used as such.
• The document does not constitute an endorsement or recommendation of any particular technology, product, or service.
• The reader assumes all responsibility for their use of the information contained in this document and any consequences that may arise.
• The author disclaim any liability for any damages or losses that may result from the use of this document or the information contained therein.
• The author and publisher reserve the right to update or change the information contained in this document at any time without prior notice.

****************************************************

Topics to be read

1. Basics of Statistics

2. Distributions

·               Binomial distribution

·              Poisson distribution

·             T- distribution

3. Regression Models

·               Linear regression

·              Logistic regression

******************************************************

Statistics

1. Statistics:

a. Is a way to get information from data.

     b. Statistics occurs only when there is Variance or Variability

2. Descriptive Statistics : There is no inference or predictions

3. Inferential Statistics : have attached numerical values

4. Population sample is denoted by Greek Alphabets σ

5. Sample Population is denoted by English Alphabets s

6. Measure of Central Tendency:

  a. mean

  b. median

  c. mode

7. Median: The one value of X, when data is arranged in array. The mid-point which  divides the population into two equal halves.

  a. Weighted Mean

8. Skewed Distribution

  a. Right skewed

  b. Left skewed

9.Kurtois: Degree to which a distribution is peaked.

10. Central Limit Theorem will convert any Non-Normal distribution into Normal distribution.

11. Quartile

There are 3 Quartile’s present

Q1

Q2

Q3

Inter Quartile Range = Q3- Q1

12. Coefficient of Variance = Sigma / Mu

13. Sum of Squared Deviations =  (X-Xbar)²,

14. Statistical Methods

     a. Linear regression

     b. Logistic regression

15. Linear regression Y = mX + C

16. Types of Variables

·       Qualitative

·       Quantitative

o    Discrete

o   Continuous

17. Binomial means : Two outcomes – 1 0r 0 ; True or False.

18. Permutations and combinations

·       Permutations means arrangement

·       How many ways you can arrange “r” from “n” objects.

19. Probability

·       For decision making there is no probability, here it is whether YES or NO.

·       Only when there is likelihood of Outcome probability occurs.

20. (A and B)  = Joint occurrence. Event A must occur and event B must occur

·       (A or B) = Any event can occur or both event can occur  

21. Addition Theorem : P(A+B) = P(A) +P(B) - P(A-B)

22. Multiplication Theorem: P(AB)= P(A) * P(B/A)

·       P(B/A) = Probability B given A.

·       Given P(A) has happened.

23. Probability

·       Classical approach: When the outcomes are known and equally distributed.

o   Example Throw of Dices.

·       Frequency approach: Using Survey and Experiment

o   To know the mileage of  cars.

·       Subjective approach:

o   Like Man landing on moon: - before probability was zero.

24. Probability Tree

25. Bayes’ Theorem:

Changing the  probability value of the event Leads to Bayes’ theorem.

26. Binomial Distribution

x-axis is Variable

y-axis is Probability

Parameters of Binomial are “n” & “p”.

27. Normal distribution

·       Bell Curve

·       Mean = Medium = Mode at Peak

·       Asymptotic -  the curve doesn’t touches the x-axis on both the ends

·       Symmetric

·       Sum under the Curve is 1.

·       Std. Normal Curve :- z=(x-Mu)/sigma

28. Central Limit Theorem:

·       When you  plot the X-bar Mean of Sampling it is always a normal distribution.

·       Recommended to take 30 samples to make a data set  into normal distribution

29. Hypothesis

·       Hypothesis means assumptions

·       H₀  (equal to )Mu₀

·       H₁  (not equal to) Mu₁

·       When evidence is  strong you reject H₀

·       When evidence is weak you are unable to reject H₀

   2-Tail test of Hypothesis 

30.  Type 1 Error and Type 2 Error.

                  Type 1 error = Alpha error

                  Type 2 error = beta error

	Guilty	Not Guilty
Guilty	yes	x	Type 2 error
Not Guilty	x	yes
	Type 1 error

Type 1 error has to be minimised

Type 1 error alpha = 0.05% allowed in model

31.  Table Distributions

Z -  table distribution = large populations

T - Table distribution = unknown populations

32. ANOVA = ANalysis Of  VAriance

Two Ways ANOVA :-

                  Example:

Type of car Brands : brand 1, brand 2, brand 3, brand 4.

Here the these are called as LEVELS = 4

                 

The parameters against which these are compared = Battery, Mileage

Here these two are called as FACTORS = 2.

Now There are 2 factors and 4 levels.  Since there are 2 factors it is called by  TWO Way ANOVA.

33.  F- Distribution

                  SSB         = Sum of square b/w columns            

                  SSW        = Sum of squares within columns

                  MSB        = Mean square b/w columns

                  MSW         = Mean square within columns

        dF            = degree of freedom

MSB = SSB/ dF

MSW = SSW/dF

(F – distribution) F = MSB/MSW

******************************************************

Regression 

34. Simple linear regression

                  Linear means a straight line

Y = m X + c

m = slope

Y = Dependent 

c = intercept

#