class: center, middle, inverse, title-slide .title[ # Welcome to Tools for Analytics ] .author[ ### Lars Relund Nielsen ] --- layout: true <div class="my-footer"> <span> <a href="https://bss-osca.github.io/tfa/index.html" target="_blank">Notes</a> | <a href="01_intro-slides.html" target="_blank">Slides</a> | <a href="https://github.com/bss-osca/tfa/blob/master/slides/01_intro-slides.Rmd" target="_blank">Source</a> </span> </div> --- ## About Me - Lars Relund Nielsen (larsrn@econ.au.dk) - Professor at Dept. of Economics and Business Economics, Aarhus Universitet. - I do research in algorithms for optimization, multiobjective optimization, stochastic dynamic programming - Programming in all kinds of languages such as C, C++, VBA, R, SQL, Javascript … - For more information see my [research page]( http://www.research.relund.dk/). --- ## Who are you? Let us have a look at [what you answered](https://docs.google.com/forms/d/1QFd_tR_LIWTNpvYHAow19DeFKa_ht56rKD6ICjLeNlY/edit#responses). <img src="./img/responses.png" width="120%" style="display: block; margin: auto;" /> --- ## Course overview - Business Analytics (BA) refers to the scientific process of transforming data into insight for making better decisions in business. - There is an increasing need for analysts with the ability to do tasks within Analytics. - This requires for the analyst to be particularly qualified in handling IT based tools. - The use of programming is the basis for performing analytical tasks such as descriptive, predictive and prescriptive analytics. - This course gives you an introduction to different tools to perform Analytics and focus on descriptive analytics. - The course is partitioned into two parts. * The VBA part gives you an introduction to the compiled programming language VBA (Visual Basic for Applications) in Excel. * The R part gives you an introduction to the interpreted programming language R. --- ## Analytics as a data driven process .pull-left[ - First you must __import__ your data. <!-- That is, take data from a database, file, web API etc. and transform it into a data frame/table. --> - Next you must get your data in the right form (__wrangling__). <!-- - In general raw data may be messy and need to be structured in a __tidy__ way. --> <!-- Tidying your data means storing it in a structured form sutaiable for analysis. In brief, when your data is __tidy__, each column is a variable, and each row is an observation. Tidy data is important because the consistent structure lets you focus your struggle on questions about the data. --> <!-- - Once you have tidy data, a common first step is to __transform__ it. --> <!-- Transformation includes narrowing in on observations of interest (e.g. only observations from a specific year or warehouse), creating new variables based on existing variables (like the cost of using the machine that day given idle time). --> - Exploration, visualization and modeling may be seen as different steps which can be used for __analyzing__ the data and answer the overall questions. <!-- This part of the course will focus on the different steps except modeling. --> <!-- - The next step is to do a simple __exploration__ of you data such as calculating a set of summary statistics (like counts, means or KPIs). --> <!-- A good way to get an overview over your data is by __visualisation__. --> <!-- A good visualisation will show you things that you did not expect, raise new questions about the data or confirm your hypotesis. A good visualisation might also hint that you're asking the wrong question, or you need to collect different data. Exploration and visusalization are descriptive analytics and used to answer questions such as: What happend? How many, how often, where? Where exactly is the problem? What actions are needed? --> <!-- - __Models__ are complementary tools to visualisation. Once you have made your questions sufficiently precise, you can use a model to answer them. A model is a description of a system using mathematical concepts and a simplification of the real system. That is, the results of a model are based on a set of assumptions. Models for statistical analysis, forecasting, system behaivior are predictive analytics and answer questions like: Why is this happening? What if these trends continue? What will happen next? Models for prescriptive analytics use optimization and other decision modelling techniques to suggest decision options with the goal of improving business performance and answer questions like: What is the best that can happen? --> - Given an analysis you must __communicate__ the results to decision makers. <!-- Note that analytics is not a one-way process, is is common that you several times have to tidy and transform your data, explore and visualize based on the results of a model, rerun the model given feedback on you communication to the decision makers etc. Common connections are visualized using directed arrows in Figure \@ref(fig:process). --> - Surrounding the process is __programming__ (your Swiss army knife). <!-- An introduction to programming is given in the first part of the course using VBA in Excel. We will cover programming using R in this part of the course. --> ] .pull-right[ <img src="./img/process.png" width="120%" style="display: block; margin: auto;" /> ] --- ## VBA Part - VBA is a programming language ”behind” MS Office. - Automate processes and create algorithms. - Introduction to programming (i.e. coding). - Advanced datatypes: arrays, cells, worksheets, workbooks ... - Input/Output, variables, loops, conditions, procedures. <!--<div class="centerVH"> <h2>Over to Lars ...<h2/> <div/>--> --- ## R Part - R basics - Programming in R (loops, conditionals and functions) - Rmarkdown - IO, tidy and transform - Visualization --- ## Teaching form and style - A combination of lectures and study cafés. - Read material, do interactive tutorials and start with exercises. - Lecture. - Continue with exercises (group work). - Join the Study café if you need help with the exercises. - The course is a "learning by doing" course. Programming is best learned by trying it yourself given some exercises. - The main focus is on giving you a set of exercises that you can solve in your study groups. - It is strongly recommended that you solve the exercises for each week during the course. This is the best way to prepare for the exam. --- ## Course notes The course notes are given [online](https://bss-osca.github.io/tfa/). Annotate the notes and slides using [hypothes.is](https://bss-osca.github.io/tfa/annotate.html). Link to slides are given under the **Recap** section. <img src="./img/online-site.png" width="100%" style="display: block; margin: auto;" /> --- ## Groups - I have formed study groups of 3-6 persons which will be used in the whole course. - It is up to you how to meet in these groups (on-line or physical). - The groups will be the backbone when you solve exercises. - It is expected that you have done some effort in understanding/solving the exercises before you meet in your group. - The project report also have to be handed in as groups. --- ## Project report - One report written during the course. - You may choose either an R or a VBA assignment. - Make the report in your group. - Do individual peer grading using the an on-line platform. - Purpose: to work with a larger assignment in R or VBA. - The assignment will be more extensive than the exam. - Do not expect such a large assignment at the exam! --- ## Exam - A three-hour written exam about all the curriculum/topics in the course. - The exam will be using either R or VBA (not both). - It will not be reviled if R or VBA is used before the exam! - Curriculum include all material such as book chapters, course notes, videos, interactive tutorials, exercises and project reports. - We will do previous exam assignments during the course, so you have a feel about the difficulty of the exam. --- layout: false class: inverse, middle, center # Happy coding during the course!