Professional Documents
Culture Documents
BAMS (Business Analytics) PPT
BAMS (Business Analytics) PPT
By –
TANISHK SINGH
Decision Analysis
Decision Analysis is a systematic and quantitative approach to decision-making that involves assessing
and analyzing various alternatives to make informed choices. It is particularly useful in complex and
uncertain situations where multiple factors need to be considered.
Key Components
• Objectives: Clearly define the objectives that the decision aims to achieve. These objectives provide a basis for
evaluating and comparing alternatives.
• Constraints: Identify any constraints or limitations that must be considered in the decision-making process.
Identification of Alternatives: Generate a comprehensive list of possible alternatives or courses of action that could address
the decision problem.
Probability Assessment:
• Assign probabilities to the uncertainties identified during the analysis. This involves estimating the likelihood of various
scenarios or outcomes.
Decision Models:
• Develop mathematical models, simulations, or other analytical tools to represent the relationships between decision
variables, uncertainties, and outcomes.
Preference Elicitation:
• Capture the preferences and priorities of decision-makers regarding the criteria and objectives.
• Use techniques such as utility functions to quantify preferences and trade-offs.
Sensitivity Analysis:
• Assess the sensitivity of the decision to changes in assumptions, input parameters, or uncertainties.
• Understand how variations in key factors affect the overall decision.
Decision Strategy:
• Develop decision strategies based on the analysis and preferences of decision-makers.
• Determine the optimal or preferred alternative considering the objectives and constraints.
Decision Tree
A Decision Tree is a graphical representation of a decision-making process that shows the possible
outcomes of different choices. It is a powerful tool in decision analysis and can be used for both
quantitative and qualitative decision-making. Decision trees are particularly useful when there are
multiple alternatives, uncertainties, and trade-offs involved in a decision.
Key Components
Decision Nodes:
• Represent decision points in the decision tree where a decision-maker must choose between alternative courses of
action. These nodes are typically depicted as squares or rectangles.
Branches:
• Connect decision nodes, chance nodes, and terminal nodes with branches. Each branch represents a possible
alternative or outcome associated with a decision or chance node.
Probabilities:
• Assign probabilities to the branches emanating from chance nodes. These probabilities indicate the likelihood of each
possible outcome occurring.
Payoffs (Utilities):
• Assign payoffs or utilities to the outcome nodes. Payoffs represent the benefits or values associated with each possible
outcome.
Decision Rules:
• Define decision rules that guide the decision-maker at decision nodes. These rules are based on the preferences, priorities,
or criteria established for the decision.
Tree Structure:
• Organize the decision tree in a hierarchical structure with decision nodes at the top, chance nodes in the middle, and
terminal nodes at the bottom. The structure visually represents the decision-making process.
Decision Strategy:
• Determine the optimal decision strategy based on the analysis. This could involve selecting the alternative with the highest
expected value or using other decision criteria.
Multiple Regression Analysis
Multiple Regression Analysis is a statistical technique used to examine the relationship between a dependent
variable and two or more independent variables. It extends the principles of simple linear regression, where
there is only one independent variable. Multiple Regression Analysis allows for a more comprehensive
understanding of how several factors may jointly influence the variation in the dependent variable.
Key Components
Intercept:
• The intercept represents the estimated value of the dependent variable when all independent variables are zero.
Coefficients:
• These coefficients represent the estimated change in the dependent variable for a one-unit change in the corresponding
independent variable, holding other variables constant.
Residuals (ε):
• Residuals are the differences between the observed values of the dependent variable and the values predicted by the
regression model. Residual analysis is important for assessing the model's goodness of fit.
Hypothesis Testing:
• Hypothesis tests are conducted to assess the significance of each coefficient. The null hypothesis typically states that the
coefficient is equal to zero.
Multicollinearity:
• Multicollinearity occurs when independent variables in the model are highly correlated. It can affect the stability and
interpretation of the coefficients.
Adjusted R-squared:
• Adjusted R(2) accounts for the number of predictors in the model and provides a more accurate measure of goodness of fit
when comparing models with different numbers of variables.
F-statistic:
• The F-statistic tests the overall significance of the regression model. It evaluates whether at least one independent
variable significantly contributes to explaining the variation in the dependent variable.
Durbin-Watson Statistic:
• It assesses the presence of autocorrelation in the residuals. A value around 2 suggests no autocorrelation.
Time Series Analysis
Time Series Analysis is a statistical technique used to analyze and interpret temporal data. Time series data consists of
observations on a variable or several variables over time. This analysis allows for the identification of patterns, trends,
and seasonality within the data.
Key Components
Temporal Patterns:
• Identify and analyze patterns in the data, including trends, seasonality, and cyclic behavior.
Model Validation:
• Validate the model by comparing predicted values to actual values in a validation dataset not used during model
training.
Outlier Detection:
• Identify and handle outliers that can significantly impact the analysis and modeling of the time series.
Smoothing Techniques:
• Apply moving averages or exponential smoothing to reduce noise in the data and highlight underlying trends or patterns.
Stationarity:
• Check for stationarity, ensuring that the statistical properties of the time series (mean, variance, and autocorrelation) do
not change over time. Differencing may be applied to achieve stationarity.