Hi everyone   I have a dataset containing data on a weekly basis. Each week has its own (character) variable - e.g. week 32 of 2010 is labelled y1032. What I want to do is to keep only data from 2018, 2019, and 2020 (keeping all weeks within a given year) - i.e keppeing 52*3 = 156 weeks:   - 1801, 1802, ... 1852. - 1901, 1902, ... 1952 - 2001, 2002, ... 2052   I think an array should do the job. However, I'm having a hard time setting up a proper array as there is a "0" before the first 9 weeks within every year. Any suggestions on how to go over this? Best, Frederik     ... View more

Hi Everyone   I have two datasets which I aim to merge.   First dataset looks like this (some of it):   "HomeTeam","AwayTeam","Date","B365H","B365D","B365A","p_home_win","p_draw","p_away_win" "Man City","Blackpool","01/01/2011","1.29","5.5","10","56.170","24.346","19.484" "Stoke","Everton","01/01/2011","2.63","3.2","2.75","21.027","27.827","51.146" "Sunderland","Blackburn","01/01/2011","1.83","3.5","4.5","60.521","24.321","15.158" "Fulham","West Brom","01/04/2011","2.2","3.25","3.4","24.902","23.708","51.390" "Aston Villa","Sunderland","01/05/2011","2.2","3.25","3.4","69.424","16.212","14.353" "Newcastle","West Ham","01/05/2011","1.83","3.6","4.33","58.885","18.826","22.284" "Wolves","Chelsea","01/05/2011","8","4.2","1.44","66.943","16.625","16.419" "Chelsea","Blackburn","01/15/2011","1.29","5.5","11","89.759","6.747","3.281" "Man City","Wolves","01/15/2011","1.33","5","10","45.785","27.370","26.845" "Stoke","Bolton","01/15/2011","2.1","3.25","3.75","56.967","24.800","18.233" "West Brom","Blackpool","01/15/2011","1.8","3.6","4.5","59.517","18.727","21.751" "West Ham","Arsenal","01/15/2011","6","4","1.57","6.850","15.515","77.632" "Wigan","Fulham","01/15/2011","2.5","3.25","2.88","20.793","27.330","51.877" "Birmingham","Aston Villa","01/16/2011","2.4","3.2","3.1","23.311","29.941","46.747" "Liverpool","Everton","01/16/2011","2.2","3.2","3.5","64.537","23.468","11.994" "Sunderland","Newcastle","01/16/2011","2.2","3.25","3.4","44.228","33.868","21.903" "Arsenal","Wigan","01/22/2011","1.2","6.5","15","49.116","23.552","27.332" "Everton","West Ham","01/22/2011","1.5","4","7.5","52.747","25.312","21.941" "Fulham","Stoke","01/22/2011","2.2","3.2","3.5","71.301","20.783","7.916" "Man United","Birmingham","01/22/2011","1.2","6.5","15","63.496","23.443","13.062" "Newcastle","Tottenham","01/22/2011","3.5","3.4","2.1","33.698","26.365","39.936" "Wolves","Liverpool","01/22/2011","3.4","3.2","2.25","9.437","15.631","74.927" "Blackburn","West Brom","01/23/2011","2.25","3.3","3.25","53.162","29.265","17.574" "Bolton","Chelsea","01/24/2011","6","3.75","1.62","29.028","23.656","47.317" "Wigan","Aston Villa","01/25/2011","3.2","3.25","2.3","53.062","23.313","23.625" "Liverpool","Fulham","01/26/2011","1.53","3.8","7","73.975","16.967","9.056" "Arsenal","Everton","02/01/2011","1.4","4.5","8.5","51.904","21.570","26.525" "Man United","Aston Villa","02/01/2011","1.33","5","9.5","67.411","23.158","9.431" "Sunderland","Chelsea","02/01/2011","5.5","3.6","1.67","19.029","33.295","47.675"     Second dataset looks like this (some of it):   "HomeTeam","AwayTeam","Date","B365H","B365D","B365A","HomeTeam_hotness_beforegame","AwayTeam_hotness_beforegame" "Arsenal","Chelsea","12/27/2010","2.63","3.2","2.75","3.7604757667","-3.760475767" "Birmingham","Man_United","12/28/2010","6.5","4.33","1.5","-0.63663219","0.6366321898" "Man_City","Aston_Villa","12/28/2010","1.44","4.2","8","1.6836796735","-1.683679673" "Stoke","Fulham","12/28/2010","1.83","3.4","4.75","2.9566006505","-2.956600651" "Sunderland","Blackpool","12/28/2010","1.62","3.75","6","-0.723181362","0.7231813625" "Tottenham","Newcastle","12/28/2010","1.53","4","6.5","2.2847551983","-2.284755198" "West_Brom","Blackburn","12/28/2010","1.83","3.5","4.5","0.382658028","-0.382658028" "West_Ham","Everton","12/28/2010","3.4","3.2","2.25","1.7574369645","-1.757436964" "Chelsea","Bolton","12/29/2010","1.29","5.5","11","-3.928073866","3.9280738659" "Liverpool","Wolves","12/29/2010","1.33","5","10","-0.822763551","0.8227635506" "Wigan","Arsenal","12/29/2010","6.5","4.2","1.5","-0.538853648","0.5388536475" "Birmingham","Arsenal","01/01/2011","5.5","3.6","1.67","-0.181423191","0.1814231907" "Liverpool","Bolton","01/01/2011","1.7","3.75","5","-1.100856462","1.1008564622" "Man_City","Blackpool","01/01/2011","1.29","5.5","10","-1.022399727","1.0223997272" "Stoke","Everton","01/01/2011","2.63","3.2","2.75","1.0295768053","-1.029576805" "Sunderland","Blackburn","01/01/2011","1.83","3.5","4.5","-0.249932422","0.249932422" "Tottenham","Fulham","01/01/2011","1.4","4.5","8.5","2.4818040774","-2.481804077" "West_Brom","Man_United","01/01/2011","6","4.2","1.53","-0.026691257","0.026691257" "West_Ham","Wolves","01/01/2011","2","3.4","3.8","-0.774912117","0.7749121174" "Chelsea","Aston_Villa","01/02/2011","1.25","5.5","13","-1.730987424","1.7309874237" "Wigan","Newcastle","01/02/2011","2.5","3.2","2.9","0.264612696","-0.264612696" "Blackpool","Birmingham","01/04/2011","2.3","3.25","3.25","2.1186630801","-2.11866308" "Fulham","West_Brom","01/04/2011","2.2","3.25","3.4","-1.037093869","1.0370938694" "Man_United","Stoke","01/04/2011","1.25","5.75","12","1.3149114291","-1.314911429" "Arsenal","Man_City","01/05/2011","1.91","3.6","4","-0.441275477","0.4412754771" "Aston_Villa","Sunderland","01/05/2011","2.2","3.25","3.4","-0.898485414","0.8984854145" "Blackburn","Liverpool","01/05/2011","3.4","3.25","2.2","0.0970333453","-0.097033345" "Bolton","Wigan","01/05/2011","1.73","3.6","5","0.8621463159","-0.862146316" "Everton","Tottenham","01/05/2011","2.63","3.25","2.75","-2.335397536","2.3353975359" "Newcastle","West_Ham","01/05/2011","1.83","3.6","4.33","0.1260786995","-0.1260787" "Wolves","Chelsea","01/05/2011","8","4.2","1.44","3.7990423986","-3.799042399" "Blackpool","Liverpool","01/12/2011","3.8","3.75","1.91","2.705250589","-2.705250589" "Chelsea","Blackburn","01/15/2011","1.29","5.5","11","-2.77646534","2.7764653398" "Man_City","Wolves","01/15/2011","1.33","5","10","-0.716695062","0.7166950617" "Stoke","Bolton","01/15/2011","2.1","3.25","3.75","-0.184180716","0.1841807157" "West_Brom","Blackpool","01/15/2011","1.8","3.6","4.5","-2.662987004","2.6629870041" "West_Ham","Arsenal","01/15/2011","6","4","1.57","0.4998071633","-0.499807163" "Wigan","Fulham","01/15/2011","2.5","3.25","2.88","-0.658881604","0.6588816037" "Birmingham","Aston_Villa","01/16/2011","2.4","3.2","3.1","0.0981564779","-0.098156478" "Liverpool","Everton","01/16/2011","2.2","3.2","3.5","-1.220109433","1.220109433" "Sunderland","Newcastle","01/16/2011","2.2","3.25","3.4","-0.381697342","0.3816973421" "Tottenham","Man_United","01/16/2011","2.88","3.3","2.5","-0.808198231","0.808198231" "Arsenal","Wigan","01/22/2011","1.2","6.5","15","0.4748149132","-0.474814913" "Aston_Villa","Man_City","01/22/2011","3.5","3.4","2.1","-1.274314497","1.2743144968" "Blackpool","Sunderland","01/22/2011","2.5","3.25","2.88","1.0883505902","-1.08835059" "Everton","West_Ham","01/22/2011","1.5","4","7.5","-0.740684864","0.7406848636" "Fulham","Stoke","01/22/2011","2.2","3.2","3.5","-0.690559941","0.6905599411" "Man_United","Birmingham","01/22/2011","1.2","6.5","15","0.8826238873","-0.882623887" "Newcastle","Tottenham","01/22/2011","3.5","3.4","2.1","0.9706523784","-0.970652378"     I want to add the values for "HomeTeam_hotness_beforegame" variable and the  "AwayTeam_hotness_beforegame" variable to the rows in the first dataset. Doing this by date hometeam awayteam should secure that SAS puts the values for the variable on the correct observation. I've tried using this code:     proc sort data=poisson_final; by date hometeam awayteam; run; proc sort data=PL_score_1; by date hometeam awayteam; run; data results; merge poisson_final PL_score_1; by date; run;   But the output doesn't make sense - see blue box in the picture below. It is like SAS is assigning prior values for "p_home_win", "p_draw" and "p_away_win".     Any suggestions on why the merge doesn't do the job for me here?   Any help is very much appreciated!   Best, Frederik       ... View more

Hi everyone   I have a dataset of 3.257 observations on Premier League matches the last 10 seasons. It looks like this: "HomeTeam","AwayTeam","Season","Date","Hometeam_goal_expectancy","Awayteam_goal_expectancy" "Man City","Blackpool","2011","01/01/2011","3.19298245614035","1.36058252427184" "Stoke","Everton","2011","01/01/2011","0.64210526315789","1.486875" "Sunderland","Blackburn","2011","01/01/2011","1.93038596491228","1.34933333333333" "Fulham","West Brom","2011","01/04/2011","1.34095238095238","1.98110552763819" "Aston Villa","Sunderland","2011","01/05/2011","2.33229629629629","1.0790206185567" "Newcastle","West Ham","2011","01/05/2011","1.83804347826086","2.00910891089108" "Wolves","Chelsea","2011","01/05/2011","3.10180505415162","1.40392156862745" "Chelsea","Blackburn","2011","01/15/2011","4.55042345276873","1.47929203539823" "Man City","Wolves","2011","01/15/2011","3.48054237288135","0.55891402714932" "Stoke","Bolton","2011","01/15/2011","1.24107744107744","1.18506787330316" "West Brom","Blackpool","2011","01/15/2011","1.70254966887417","2.33219730941704" "West Ham","Arsenal","2011","01/15/2011","0.85485197368421","2.13094170403587" "Wigan","Fulham","2011","01/15/2011","0.15635179153094","1.67400881057268" "Liverpool","Everton","2011","01/16/2011","1.86122448979591","0.9035294117647" "Sunderland","Newcastle","2011","01/16/2011","1.014","1.85294117647058" "Arsenal","Wigan","2011","01/22/2011","1.34653465346534","1.58260869565217" "Everton","West Ham","2011","01/22/2011","2.49180327868852","0.63436123348017" "Fulham","Stoke","2011","01/22/2011","1.49837133550488","0.69868995633187" "Man United","Birmingham","2011","01/22/2011","0.97087378640776","0.78260869565217" "Newcastle","Tottenham","2011","01/22/2011","1.16504854368932","1.07079646017699" "Wolves","Liverpool","2011","01/22/2011","0.70358306188925","1.23893805309734" "Blackburn","West Brom","2011","01/23/2011","1.19736842105263","0.6695652173913" "Bolton","Chelsea","2011","01/24/2011","1.18032786885245","1.85022026431718" "Wigan","Aston Villa","2011","01/25/2011","1.29449838187702","2.21973094170403" "Liverpool","Fulham","2011","01/26/2011","3","1.14285714285714" "Arsenal","Everton","2011","02/01/2011","2.75496688741721","0.76018099547511" "Man United","Aston Villa","2011","02/01/2011","1.64705882352941","0.50224215246636" "Sunderland","Chelsea","2011","02/01/2011","0.27814569536423","2.48888888888888" "Birmingham","Man City","2011","02/02/2011","0.29220779220779","2.88288288288288"   For every observation I want to calculate the probability of all possible scorelines using the Poisson distribution. I restrict the team to score a maximum of 10 goals.   For instance for the first observation (see data above) I can manually plug in "3.19" and "1.36" into this code: data final_6; do Home_goals=0 to 10; do Away_goals=0 to 10; format probability 8.3; probability = pdf ('poission', Home_goals, 3.19) * pdf('Poisson', Away_goals, 1.36)*100; output; end; end; run; (…) to get this table (containing 121 observations):   I aim to avoid manually typing in the data for every game (3.257 games). Also it would be neat if the output for the probability of the different scorelines could somehow be added horisontally to the observation for the match in question. So for the first game I would end up with something like this (including all the possible scorelines): HomeTeam AwayTeam Season Date Hometeam_goal_expectancy Awayteam_goal_expectancy 0-0 1-0 2-0 3-0 4-0 ... Man City Blackpool 2011 01/01/2011 3.1929824561 1.3605825243 1.057 3.3117 5.377 5.717 I know this will add 121 variables to every observation. But I will use them to sum up the joint winning probability for the home team (simply summing alle the probabilities where the hometeam have got more goals than the awayteam) and the away team (simply summing alle the probabilities where the awayteam have got more goals than the hometeam).    Any help is very much appreciated!!   Best,   Frederik             ... View more