└── README.md


/README.md:
--------------------------------------------------------------------------------
 1 | ## Overview 
 2 | Create a model to predict the result of a game between team 1 and team 2, based on who's home and who's away, and on whether or not the game is friendly or not.
 3 | 
 4 | The two possible approaches (as  shown below) have been used to make predictions given the below inputs
 5 | - Input: Home team, Away team, Tournament type (World cup, Friendly, Other)
 6 | 
 7 | ### Approach 1: Polynomial approach
 8 | The model should use the below features:
 9 | - Rank of home team
10 | - Rank of away team
11 | - Tournament type
12 | 
13 | Model 1: Predict how many goals the home team scores
14 | Model 2: Predict how many goals the away team scores
15 | 
16 | ### Approach 2 Logistic approach
17 | Feature Engineering: Figure out from the home team’s perspective if the game is a Win, Lose or Draw (W, L, D)
18 | 
19 | ### Context
20 | A more detailed explanation and history of the rankings is available [here](https://en.wikipedia.org/wiki/FIFA_World_Rankings)
21 | 
22 | An explanation of the ranking procedure is available [here](https://www.fifa.com/fifa-world-ranking/procedure/men.html)
23 | 
24 | ### Dataset Columns
25 | Some features are available on the FIFA [ranking page](https://www.fifa.com/fifa-world-ranking/ranking-table/men/index.html)
26 | 
27 | - Rank
28 | - Country Abbreviation
29 | - Total Points
30 | - Previous Points
31 | - Rank Change
32 | - Average Previous Years Points
33 | - Average Previous Years Points Weighted (50%)
34 | - Average 2 Years Ago Points
35 | - Average 2 Years Ago Points Weighted (30%)
36 | - Average 3 Years Ago Points
37 | - Average 3 Years Ago Points Weighted (20%)
38 | - Confederation
39 | - Date - date of the match
40 | - Home_team - the name of the home team
41 | - Away_team - the name of the away team
42 | - Home_score - full-time home team score including extra time, not including penalty-shootouts
43 | - Away_score - full-time away team score including extra time, not including penalty-shootouts
44 | - Tournament - the name of the tournament
45 | - City - the name of the city/town/administrative unit where the match was played
46 | - Country - the name of the country where the match was played
47 | - Neutral - TRUE/FALSE column indicating whether the match was played at a neutral venue
48 | 
49 | ### Deliverables
50 | - Perform  Exploratory Data Analysis
51 | - Perform any necessary feature engineering
52 | - Check of multicollinearity
53 | - Build models
54 | - Cross-validate the models
55 | - Compute RMSE
56 | - Create residual plots for the models, and assess their heteroscedasticity using Bartlett’s test
57 | 
58 | ### Dataset
59 | The dataset and glossary to use for this project can be found [here](https://drive.google.com/open?id=1BYUqaEEnFtAe5lvzJh9lpVpR2MAvERUc)
60 | 
61 | 


--------------------------------------------------------------------------------