Topic1_Part2_Introduction_2023_2.pptx

Topic 1: Introduction GEN 4191 Data Analytics for Business Optimisation 2023.2 – BBA6 Dr. Krisztina Soreg What is Business Analytics? The use of data, information technology, statistical analysis, quantitative methods and mathematical methods to help managers to make better (fact-based) decisio...

Topic 1: Introduction GEN 4191 Data Analytics for Business Optimisation 2023.2 – BBA6 Dr. Krisztina Soreg What is Business Analytics? The use of data, information technology, statistical analysis, quantitative methods and mathematical methods to help managers to make better (fact-based) decisions. Fact: everyone makes decisions Problem 1: too much available data Problem 2: unreliable data & unpredictable future Common types of decisions: • • • • Pricing Customer segmentation Merchandising Location More info: Textbook Chapter 1 Four types of analytics 4. Prescriptive: using highly advanced tools and techniques to assess the consequences of possible decisions and determine the best course of action in a scenario 1. Descriptive: to show what has already happened in a business (e.g.: averages and percent changes) 3. Predictive: using findings from descriptive and diagnostic analytics – along with sophisticated predictive modeling, machine learning and deep learning techniques – to predict what will happen next. 2. Diagnostic: to identify the root causes of events and behaviors Main types of Analytics Descriptive Analytics • use of data to understand past and current business performance and make informed decisions. • the most commonly used analytics • summarizes data into meaningful charts and reports: budgets, sales, revenues or cost • goal: to obtain standard and customized reports • e.g.: impact of an advertising campaign review business performance to find problems or areas of opportunity, and identify patterns and trends in data • Example: classify customers into different segments, which enables them to develop specific marketing campaigns Main types of analytics Predictive Analytics • seeks to predict the future by examining historical data, detecting patterns or relationships in these data and then extrapolating these relationships forward in time • detecting hidden patterns in large quantities of data to segment and group data into coherent sets to predict behavior and detect trends • e.g.: a bank manager might want to identify the most profitable clients or predict the chances that a loan applicant will default • predicting the response of different customer segments to an advertising campaign Main types of analytics Prescriptive Analytics Using optimization to identify the best alternatives to minimize or maximize some objectives. • Problem: too many choices or alternatives for a human decision maker • e.g.: best pricing and advertising strategy to maximize revenue, the optimal amount of cash to store in ATMs, or the best mix of investments in a retirement portfolio to manage risk Main types of analytics Find out more about Data Analytics with examples by reading this article: https://www.sap.com/insights/what-is-analytics.html What is Statistics? What is Statistics? The science concerned with developing and studying methods for collecting, analyzing, interpreting and presenting empirical data  the study of uncertainty More info: Textbook Chapter 1 What is Statistics? Descriptive Types of Statistics Inferential What is Statistics? Descriptive Statistics • The meaningful presentation of data such that its characteristics can be effectively observed. • Used to describe the basic features of the data in a study. They provide simple summaries about the sample and the measures. Together with simple graphics analysis, they form the basis of virtually every quantitative analysis of data. 1. Graphical display 2. Tables: condensation of large masses of data into a form that is more readily understood 3. Preparation of summary measures to give a concise description of the information: Average / Mode. What is Descriptive Statistics? Method: quantitatively describing the important characteristics of the dataset What is Inferential Statistics? Inferential Statistics • Relates to decision making. • We try to reach conclusions that extend beyond the immediate data alone  generalizing from the sample to the population: probability, testing hypotheses, etc. • Which decisions? 1. Determining whether any apparent characteristics of situation are unusual. 2. Estimating the value of unknown numerical quantities and determining the reliability of those estimates. 3. Using past occurrences to attempt to predict the future. What is Inferential Statistics? Inferential Statistics - Example Question: which factors might have an impact on the decline of poverty in a country? Possible factors: - GDP per capita - Number of people with college / university degree - Unemployment rate - Number of medical personnel - Availability of 5G - etc. Method: to predict the relationship between variables  which of them is having a significant impact in our research (regression analysis) What is Inferential Statistics? What do we see? 1) Higher the GDP per capita  higher national poverty line 2) Higher national poverty line  higher GDP per capita What is for sure: there is a strong, statistical link between economic growth (development) and poverty reduction Population & Sample Population Our goal: to make a statement, observation about the entire population Problem: almost impossible to access the full population! Sample Solution: generalizing from a small data set originating from the population E.g.: a sample of 1,000 citizens is taken from the population of all Canadian citizens Key Terms Metrics (scales) • A unit of measurement that provides a way to objectively quantify performance • E.g.: senior managers might assess overall business performance using such metrics as net profit, return on investment, market share, and customer satisfaction Ratio data Nominal data Ordinal data Interval data Key Terms: Types of Metrics 1) Nominal (categorical) data • A variable determined by categories. The variables are given a descriptive name or label to represent their value. • The categories bear no quantitative relationship to one another  calculation is not possible: can’t be multiplied, divided, added or subtracted, no difference between data points! • E.g.: geographical location of clients, hair colour, gender, postal code, political party, religion, etc. 2) Ordinal data • Data can be compared to one another  order matters but not the difference between values, can’t be added to or subtracted from! • No fixed units of measurement: extent or limit of something • E.g.: rating a service as poor, average, good, very good or excellent, income level, education level, satisfaction rating, etc. Key Terms: Types of Metrics 3) Interval data • Constant differences between observations and have arbitrary zero points (we can go below 0)  no starting point or true zero (negative numbers are possible)  can be added and subtracted only • Values measured along a scale (computable), with each point placed at an equal distance from one another: order + difference • E.g.: temperature (Farenheit), temperature (Celcius), pH, credit score 4) Ratio data • Continuous data having a natural (true) zero (an object is twice as big or as long as another)  category, order, difference • It can be added, subtracted, divided and multiplied • True zero: the data has no value point (you can’t have 0 kilos) • There can be no negative variable: we cannot go below 0! • E.g.: Kelvin scale (temperature), height, speed, dollars, profit Key Terms: Types of Metrics Types of metrics: Let’s practice! Nominal Ordinal Interval Length of desks in an office X Socio-economic status of citizens (low, middle, high income level) Zip code of employees Customer satisfaction with the room booking app IQ test of employees in scores Monthly net salary of a full-time employee Ratio X X X X X 2 types of quantitative data • A whole number that can’t be divided or broken into individual parts, fractions or decimals • Countable in a finite amount of time • Best visual tool: bar & pie charts • E.g.: number of students, population of a country, number of employees in a firm, days of the months (limited) Discrete data • Values that can be broken down into different parts, units, fractions and decimals • Measurable: continuous data points  not countable! • Best visual tool: histogram • E.g.: height, weight, time, temperature, width, speed of cars, distance, length of a film Continuous data Let’s sum it up! Thank you for your attention! [email protected]