Compacting the code by using an for loop or maybe an struct?

Hello,

I am an first year student working with Matlab (so little experience)

For an assignment i need to display the min, max and median of 4 1x25 array's (as you can see below)

The question is: Is there a way in which i can compact or change the code, by for example the usage of for-if statements or structs?

[Jan]

Itwould be much easier to suggest a solution, if you post the code as text, such that we can reuse it by copy&paste.

[Tygo Nikamp]

I´m sorry, should´ve thought of that. Annyway, here´s the code:

%Loop settings
CLate = 0;       %Empty variable used as a counter in a for loop for late movement
CEarly = 0;       %Empty variable used as a counter in a for loop

for i = 1 : TotalParticipants
   if Participant(i).GKMovement == 0        %If true the goalkeeper moved late
       CLate = CLate + 1;                   %Counter + 1      
       GKLateKickDep(CLate) = Participant(i).KicksDependent;         %Creates an array with the filtered data: Late GK movement and a dependent kick strategy
       GKLateKickIndep(CLate) = Participant(i).KicksIndependent;     %Creates an array with the filtered data: Late GK movement and a independent kick strategy
   else                                     %The GK didn't move late so what remains is early GK movement
       CEarly = CEarly + 1;                 %Counter + 1
       GKEarlyKickDep(CEarly) = Participant(i).KicksDependent;       %Creates an array with the filtered data: Early GK movement and a dependent kick strategy
       GKEarlyKickIndep(CEarly) = Participant(i).KicksIndependent;   %Creates an array with the filtered data: Early GK movement and a independent kick strategy
   end
end

%Calculating the min-max and mean values of each of the here above created categories to be displayed later on.     
MaxGKLateKickDep = max(GKLateKickDep);
MinGKLateKickDep = min(GKLateKickDep);
MedianGKLateKickDep = median(GKLateKickDep);

MaxGKLateKickIndep = max(GKLateKickIndep);
MinGKLateKickIndep = min(GKLateKickIndep);
MedianGKLateKickIndep = median(GKLateKickIndep);

MaxGKEarlyKickDep = max(GKEarlyKickDep);
MinGKEarlyKickDep = min(GKEarlyKickDep);
MedianGKEarlyKickDep = median(GKEarlyKickDep);

MaxGKEarlyKickIndep = max(GKEarlyKickIndep);
MinGKEarlyKickIndep = min(GKEarlyKickIndep);
MedianGKEarlyKickIndep = median(GKEarlyKickIndep);

Answer 1

[Jan]

You do not need the initial loop. Easier:

moved            = ([Participant.GKMovement] == 0);
GKLateKickDep    = Participant(moved).KicksDependent;
GKLateKickIndep  = Participant(moved).KicksIndependent;
GKEarlyKickDep   = Participant(~moved).KicksDependent;
GKEarlyKickIndep = Participant(~moved).KicksIndependent;

But a main problem is still the pile of variables, which care important information in the name. It is better to store data in the values.

GK(1).moved     = true;
GK(1).Dependent = true;
GK(1).Value     = Participant(moved).KicksDependent;
... etc.    

Then you can run a loop over the elements of GK, which calculates the statistics.

[Tygo Nikamp]

Thanks a lot!

[Tygo Nikamp]

I have been trying it out for a few hours but i just cant get ik working. The first pat that is supposed to replace the loop only delivers me the first of 25 values i need, I have tried looping it but it would only work for the GKLateKickDep & GKLateKickIndep.

Loop code:

for i = 1: 25

moved = ([Participant.GKMovement] == 0);

GKLateKickDep(i) = Participant(moved+i).KicksDependent

GKLateKickIndep(i) = Participant(moved+i).KicksIndependent

GKEarlyKickDep(i) = Participant(~moved+i).KicksDependent

GKEarlyKickIndep(i) = Participant(~moved+i).KicksIndependent

end

I have added a picture of the excel sheet to try to explain which data i want splitted. The data is later on boxplotted so i need it in an array/ 1x25 double (at least i think)

Thanks in advance

[Jan]

No, you do not need a loop anymore. [Participant.GKMovement] == 0 is a logical index already. Using this you can omit the loop.

[Tygo Nikamp]

Could you explain how? Because i tried the logical loop and it doesn't create the arrays/data i need for the boxplot or for calculating min/max/median.

[Jan]

I cannot know, which data you need for your boxplot.

This piece of code:

moved            = ([Participant.GKMovement] == 0);
GKLateKickDep    = Participant(moved).KicksDependent;
GKLateKickIndep  = Participant(moved).KicksIndependent;
GKEarlyKickDep   = Participant(~moved).KicksDependent;
GKEarlyKickIndep = Participant(~moved).KicksIndependent;

can replace your loop:

%Loop settings
CLate = 0;       %Empty variable used as a counter in a for loop for late movement
CEarly = 0;       %Empty variable used as a counter in a for loop
for i = 1 : TotalParticipants
   if Participant(i).GKMovement == 0        %If true the goalkeeper moved late
       CLate = CLate + 1;                   %Counter + 1      
       GKLateKickDep(CLate) = Participant(i).KicksDependent;         %Creates an array with the filtered data: Late GK movement and a dependent kick strategy
       GKLateKickIndep(CLate) = Participant(i).KicksIndependent;     %Creates an array with the filtered data: Late GK movement and a independent kick strategy
   else                                     %The GK didn't move late so what remains is early GK movement
       CEarly = CEarly + 1;                 %Counter + 1
       GKEarlyKickDep(CEarly) = Participant(i).KicksDependent;       %Creates an array with the filtered data: Early GK movement and a dependent kick strategy
       GKEarlyKickIndep(CEarly) = Participant(i).KicksIndependent;   %Creates an array with the filtered data: Early GK movement and a independent kick strategy
   end
end

But the problem remains, that important information are coded in the namesof the variables. This makes it much harder to process the variables in a loop. So prefer to create an array of variables and store the information e.g. in the fields of a struct:

DepStrList = {'KicksIndependent', 'KicksDependent'};
iGK = 0;
for moved = [false, true]
   for dependent = [false, true]
       DepField      = DepStrList{dependent + 1};
       iGK           = iGK + 1;
       GK(iGK).moved = moved;
       GK(iGK).dep   = dependent;
       GK(iGK).value = Participant(moved).(DepField);
   end
end

for k = 1:numnel(GK)
   GK(k).min    = min(GK(k).value);
   GK(k).max    = max(GK(k).value);
   GK(k).median = median(GK(k).value);
end

Do you see it? Now you could add a lot of further conditions easily without getting confusing names like MedianGKEarlyKickIndep.

[Tygo Nikamp]

I think i understand what you are doing here, but i cant get it working. It gives an error on the GK(iGK).value = Participant(moved).(DepField); line were its supposed to store the IndependentKicks percentage value. Here is a picture of the data i am using maybe it will help you understand. Im sorry that i'm taking up your time but i just can't quite comprehend what you are doing. Thanks in advance again!

[Jan]

Your questions are welcome. If I do not find the time to answer, I do not answer.

You mention, that you get an error. So please post the error message.

[Tygo Nikamp]

The error message is: Insufficient number of outputs from right hand side of equal sign to satisfy assignment.

Error in PenaltyAnalysisScript (line 54)

GK(iGK).value = Participant(moved).(DepField);

Answer 2

[David Hill]

a=Participant.GKMovement==0;
GK{1}=Participant.KicksDependent(a);
GK{2}=Participant.KicksIndependent(a);
GK{3}=Participant.KicksDependent(~a);
GK{4}=Participant.KicksIndependent(~a);
for k=1:4
  MaxMinMeanGK(k,:)=[max(GK{k},min(GK{k},mean(GK{k})))];
end

[Tygo Nikamp]

Thanks a lot!