Call : (+91) 968636 4243
Mail : info@EncartaLabs.com
EncartaLabs

Business Analytics

( Duration: 6 Days )

The Business Analytics training course provides skills to understand and apply analytics in the business context. Topics covered are multiple linear regression, Binary logistic regression, Decision tree, Random forest, K means clustering, sentiment analysis in text analytics and social media analytics.

By attending Business Analytics workshop, delegates will learn:

  • Advanced working knowledge in data analytics
  • To identify and build predictive models suitable to solve business scenarios and derive insights

  • Basic awareness of analytics and statistics

COURSE AGENDA

1

Descriptive Statistics

  • Data classification and data types
  • Types of analysis- Univariate, Bivariate and Multivariate
  • Measures of central tendency
    • Mean, Mode, Median
    • When to use Mean or median as a measure of central tendency
    • Percentiles, Quartiles
  • Measures of dispersion
    • Range, Interquartile Range, Mean absolute deviation, standard deviation, variance, Chebyshev’s general theorem for data distributions, Coefficient of variance
  • Shape parameters- Skewness, Kurtosis
  • Checking for Normality
    • Box & Wishker Plot
    • P-P plots
    • Q-Q plots
  • Correlation analysis and scatter plots
2

Probability Theory

  • Why use probability in business?
  • Types and methods of assigning probability
  • Marginal, Joint and Conditional probability
  • Laws of multiplication and addition of probabilities
  • The Bayes Theorem and the Bayesian school of thought
  • The concept of prior, posterior and revision of probabilities
3

Probability distributions

  • What does distributions mean to business?
  • Discrete distributions- Binomial and Poisson distributions
  • Continuous distributions- Normal distribution and the generalized exponential distribution
4

Sampling and sampling distribution

  • Why sample?
  • Types of sampling methods- Random and Non Random Sampling methods
  • Why is randomization so important- context from data modeling and market research perspective?
  • Distribution of sampling mean and the central limit theorem
  • Estimating the required sample size
5

Estimation

  • Point estimates- making inferences about sample and population parameters
  • Concept of confidence interval
6

Hypothesis testing

  • What is a hypothesis and why do we need it?
  • The concept of null hypothesis and the law of natural justice
  • Types of errors in testing
  • The power of a test
  • When to use a Single tail and when to use a two tail test?
7

Tests of Significance

  • Z –test – one tail and two tail test
  • T test- single tail, two tail and paired sample tests
  • The significance of equality of variance in a test
  • The F test for equality of variance
  • Tests for proportions
  • ANOVA – one way, two way and multiway
  • Design of experiments
8

Non-parametric Testing ( Analysis of Categorical data)

  • CHI square test for good ness of fit and independence
  • Run test for testing randomness of data
  • Man- Whitney U test for testing two samples
  • Wilcoxon Matched pair test
  • Kruskal- Walis test for Analysis of Variance
  • Friedman test for multiway analysis of variance
  • Kendal Tau test for correlation of ordinal data
  • Spearman rank correlation for nominal data
9

Data Exploration and Data preparation

  • Checking for missing values
  • Imputation techniques for missing values
  • Checking for normality of Ration data
  • Checking for distribution of the data
  • Transformation methods- logarithmic, polynomial, inverse, power transform
  • Transformation of data- when to use what
10

Data Visualization

  • Scatter plots
  • Concept of visual psychology and the gestalt philosophy
  • Fundamentals of visual elements
  • The concept of storytelling and story boarding
  • Multi-dimensional scaling and the concept of Eigen values
  • Heat maps, tree maps
  • Visualizing complex models such as based on association
11

Linear Regression

  • Simple linear regression
  • Multi linear regression
  • Regression model building
  • Concept of standardized and unstandardized coefficients
  • Regression model diagnostics and the BLUE principal
  • Role of outliers and leverage analysis
  • Step wise regression-Forward and Backward
  • Nested models and comparing models
12

Logistic Regression

  • The odds ratio and the logit transform
  • The purpose of a logit transform
  • Direct and indirect effects- the police case study
  • The effects plots
13

Cluster Analysis for segmentation and profiling

  • The concept of distances, similarities and dissimilarities
  • The popular metric distances- Euclidean, Manhattan, Mahalanobis, Miskowski
  • The popular non metric distances- Jaccard, Gower, chi square
  • The concept of linkages- average, complete, single
  • The nearest neighbor and the wards concept
  • Hierarchical clustering
  • K means clustering
  • Partioning around Mediods
  • Case study of a telecom data segmentation
  • The practical Business situation- Oh no the data is a mixed data!!!
  • Latent class clustering for mixed data
14

Decision Trees

  • The CART, CHAID and C4.5 algorithms
  • The concept of information theory and Entropy reduction
  • Business rules
  • The efficiency of a decision trees
15

Time Series Analysis

  • Assumptions of a time series
  • Components of a time series
  • Smoothing of time series - Simple moving averages and exponential smoothing
  • Seasonality and seasonal indexes
  • Time series regression analysis
  • The basic of Simple, double and Holt Winter method for forecasting
  • The random walk model
  • The concept of Auto correlation factor (ACF) and partial auto correlation factor (PACF)
  • ARIMA modeling
  • Case study of a violent pharma time series
  • Gentle introduction to -Spectral decomposition of Time series and GARCH modeling
16

Classification

  • Support Vector Machines
  • Decision Trees- CHAID, CART and C4.5 Algos
  • Naïve Bayes Classification
  • Random Forest
17

Text Mining

  • TDM/ DTM
  • Text classification
  • N gram approach to text mining
18

Monte Carlo Markov chain analysis (MCMC)

  • Simulation techniques in data mining
  • Markov chain application for brand switching

Encarta Labs Advantage

  • One Stop Corporate Training Solution Providers for over 6,000 various courses on a variety of subjects
  • All courses are delivered by Industry Veterans
  • Get jumpstarted from newbie to production ready in a matter of few days
  • Trained more than 50,000 Corporate executives across the Globe
  • All our trainings are conducted in workshop mode with more focus on hands-on sessions

View our other course offerings by visiting https://www.encartalabs.com/course-catalogue-all.php

Contact us for delivering this course as a public/open-house workshop/online training for a group of 10+ candidates.

Top
Notice
X