Final for Intro to Business Analytics

Data Overload

increasing amount of data may hinder work of analysts

data

raw facts

context

the setting, events, situation, etc

information

data organised in a way that is useful to the analyst (combines data and context)

knowledge

understanding or familiarity with information gained

decisions

conclusions reached after considering information and knowledge

business analytics

scientific process of transforming data into insight for making better decisions. Use of data to create knowledge

SOAR

• specify the question

• obtain the data

• analyse the data

• report the results

relational database

collection of tables with predefined relationships between them

table

data organised into sets of columns (fields) and rows (records)

big data

• volume

• velocity

• veracity

• variety

categorical data

words or non-meaningful numerical data— cannot perform meaningful arithmetic operations

numerical data

can perform meaningful numeric and arithmetic operations

Nominal data

categorical data that can’t be ranked

ordinal data

natural order that allows them to be ranked

interval data

equal or definitive interval between each data point; no meaningful zero

ratio data

zero is meaningful

frequency distribution

summary of data that shows the number of observations in each of several non-overlapping classes

percent frequency distribution

summarises of the percent frequency distribution of data of each bin or class

skewed left

mean < median

skewed right

mean> median

coefficient of variation

descriptive statistic that indicates how large standard deviation is relative to the mean

two-tailed test

is there a difference

one-tailed test

direction of the difference

descriptive analytics techniques

• counts

• totals, sums, averages, subtotals, cross-tabulations

• graphs, histograms, other visualisations

• vertical and horizontal analysis of financial statements

• ratio analyses

drill-down analysis

summarises data of different levels, understand why something happened

diagnostic analytics techniques

• cash/bank reconciliation

• testing for duplicates

• fuzzy matching

• sequence check

• benford’s law

benford’s law

log10(1 +1/d)

predictive analytics techniques

• regression

• classification

• forecasting with time series analysis

regression

predicts specific outcome based on independent variable inputs

forecasting with time series analysis

predicts future values based on past values of the same variable

experimental region

range of independent variables in data used to estimate the model

overfitting

results from creating an overly complex model to explain idiosyncrasies in sample data

time series

sequence of observations on a variable measured at successive points in time or over successive periods in time

trend pattern

shows gradual shifts or movements to relatively higher or lower values over a larger period of time

seasonal patterns

recovering patterns over successive periods of time

mean absolute error/deviation

average of absolute value of the forecast errors

mean absolute percent error

average of the absolute value of the percent forecast errors

mean squared error

average of the squared forecast errors

optimisation problem

• max or min some function

• set of restrictions called constraints

problem formulation

translating the verbal statement or model

• understand the problem

• define decision variables

• write objectives of decision variables

• write constraints of decision variable

answer report

• a binding constraint is one that hold an equality at the optimal solution

• slack value for each less-than-equal-to constraint indicates differences between the left and right hand values for a constraint

sensitivity analysis

the study of how changes in input parameters of an optimisation model affect the optimal solution

shadow price

change in the objective function value resulting from a one-unit change in the RHS value of the constraint

sensitivity range for an objective function value

the range of values over which the current optimal solution point will remain optimal

reduced cost

how much the objective function coefficient can be changed before the optimal solution changes

decision analysis

develop a strategy when faced with several decision alternatives and an uncertain or risk-filled pattern of future events

opportunity loss

represents the regret that people feel after making a non-optimal decision; difference between the payoff of the best decision and the payoff of the chosen decision

internal marketing analytics

• ERP for sales, transactions, and pricing data

• CRM system

• customer loyalty programs

• web analytics in company websites

external marketing analytics

• social media

• weather

• culture, laws, business environments

• various free and for-profit websites

predictive analytics techniques

• classification

• decision trees

• regression

prescriptive analytics techniques

• cash flow analysis

• goal seek analysis

• optimisation analysis

• scenario analysis

• sensitivity analysis

• sentiment analysis

accounting analytics

• financial

• managerial

• auditing

• tax

ELABORATE

financial analytics

• stock return data

• summarised financial statements

• financial statements

operations

• the part of the business organisation that is responsible for producing goods or services

• manufacturing and service processes used to transform resources into goods or services

goods

physical items such as parts, subassemblies, and final products

services

activities that provide some combination of time, location, form or psychological value

operations management

the design, operation, and improvement of the systems that create and deliver the firm’s products and/or services

supply chain

sequence of organisations— their facilities, functions, and activities— that are involved in product and delivering a product or service

supply chain management

strategic coordination of business functions within a business organisation and throughout its supply chain for the purpose of integrating supply and demand management

HR analytics

analyses human resource data to improve a company’s workforce performance

IT operations analytics

analyses company data from various applications, devices, and IT infrastructure to help managers identify, predict, solve, and prevent IT-related issues

supply chain analytics

analyses company data to help managers improve supply chain planning, efficiency, and effectiveness