Easy Notes of Data Analytics unit- 2 @Computer Diploma

Description

2.1 Graphical Techniques & Descriptive Stats

• Box Plot: Visualizes the 5-number summary (Min, Q1, Median, Q3, Max); identifies outliers.

• Skewness: Measure of asymmetry. (Left/Negative, Right/Positive).

• Kurtosis: Measure of tailedness or “peakedness” (Leptokurtic, Mesokurtic, Platykurtic).

• Descriptive Stats: Summary of data using Central Tendency (Mean, Median, Mode) and Dispersion (Range, Variance, Standard Deviation).

2.2 Correlation, Regression & Data Cleaning

• Correlation ($r$): Strength and direction of a linear relationship between two variables (-1 to +1).

• Regression: Predicting a dependent variable ($Y$) based on independent variables ($X$).

• Data Cleaning: Process of fixing corrupt, inaccurate, or formatting errors in a dataset.

2.3 Imputation Techniques

• Mean/Median Imputation: Replacing missing values with the average or middle value.

• Mode Imputation: Used for categorical missing data.

• K-NN Imputation: Replacing missing data based on the similarity of neighboring data points.

2.4 ANOVA & Chi-Square

• ANOVA (Analysis of Variance): Testing if the means of 3 or more groups are significantly different.

• Chi-Square ($\chi^2$): Testing the relationship between categorical variables (Goodness of fit or Independence).

2.5 Scatter Diagram

• Visualization: Plotting individual data points on an $X-Y$ axis.

• Pattern Recognition: Used to visually identify clusters, trends, and outliers.

2.6 Estimation & Hypothesis Testing

• Estimation: Using sample data to estimate a population parameter (Point vs. Interval).

• Null Hypothesis ($H_0$): Assumption of “no effect” or “no difference.”

• P-Value: Probability that the observed result happened by chance; compare to Alpha ($\alpha$).

2.7 Sampling Distributions & Counting

• Central Limit Theorem (CLT): As sample size increases, the distribution of the sample mean becomes normal.

• Standard Error: The standard deviation of a sampling distribution.

• Counting: Permutations (order matters) and Combinations (order doesn’t matter).

2.8 Probability & Distributions

• • Probability: Likelihood of an event occurring (0 to 1).

  • Binomial: Discrete distribution for success/failure outcomes.

  • Poisson: Discrete distribution for events occurring in a fixed interval of time/space.

  • Normal (Gaussian): Bell-shaped curve defined by mean ($\mu$) and standard deviation ($\sigma$).