Creating a grouped bar plot from a table having multiple column

I have the following table:

ValuationRatios company industry sector

___________________________________ _______ ________ ______

"P/E Ratio" 31.96 0 0

"Price/Revenue" 7.79 6.35 7.04

"Price/Book" 45.88 25.45 10.02

"Price to Cash Flow per Share" 28.46 1.33 3.89

"Price to Free Cash Flow per Share" 31.66 1.82 6.37

"Dividend Yield" 0.49 0.49 0.6

"Payout Ratio" 15.58 15.59 28.8

"Quick/Acid Test Ratio" 0.75 0.61 1.03

whos Tindclean

Name Size Bytes Class Attributes

Tindclean 23x4 4221 table

Tindclean.Properties.VariableTypes

string double double double (1 x 4) string

I am trying to create a grouped bar plot/stacked from the table.

Answer 1

[dpb]

VNames=["ValuationRatios","company","industry","sector"];
VR=[
"P/E Ratio"
"Price/Revenue"
"Price/Book"   
"Price to Cash Flow per Share" 
"Price to Free Cash Flow per Share"
"Dividend Yield"                   
"Payout Ratio"                     
"Quick/Acid Test Ratio"];            
Data=[
31.96          0          0 
7.79       6.35       7.04 
45.88      25.45      10.02 
28.46       1.33       3.89 
31.66       1.82       6.37 
0.49       0.49        0.6 
15.58      15.59       28.8 
0.75       0.61       1.03 ];
tData=[table(VR) array2table(Data)];
tData.Properties.VariableNames=VNames;
tData=convertvars(tData,'ValuationRatios','categorical')

tData = 8×4 table
             ValuationRatios             company    industry    sector
    _________________________________    _______    ________    ______

    P/E Ratio                             31.96          0          0 
    Price/Revenue                          7.79       6.35       7.04 
    Price/Book                            45.88      25.45      10.02 
    Price to Cash Flow per Share          28.46       1.33       3.89 
    Price to Free Cash Flow per Share     31.66       1.82       6.37 
    Dividend Yield                         0.49       0.49        0.6 
    Payout Ratio                          15.58      15.59       28.8 
    Quick/Acid Test Ratio                  0.75       0.61       1.03 

bar(tData.ValuationRatios,tData{:,2:end})

[Prashant joshi]

Thanks. I appreciate it. Perfect! I have follow up question, if I do not have to type variable names ( some data have several hundred variables) is there any way to tackle that?

[Prashant joshi]

it looks like there is a scaling issue. There are some variables with different scales than others and i do not want to normalize them. can you please help? my data file is attached.

Answer 2

[Cris LaPierre]

Tindclean = readtable('Tindclean.xlsx')

Tindclean = 8×4 table
               ValuationRatios               company    industry    sector
    _____________________________________    _______    ________    ______

    {'P/E Ratio'                        }     31.96          0          0 
    {'Price/Revenue'                    }      7.79       6.35       7.04 
    {'Price/Book'                       }     45.88      25.45      10.02 
    {'Price to Cash Flow per Share'     }     28.46       1.33       3.89 
    {'Price to Free Cash Flow per Share'}     31.66       1.82       6.37 
    {'Dividend Yield'                   }      0.49       0.49        0.6 
    {'Payout Ratio'                     }     15.58      15.59       28.8 
    {'Quick/Acid Test Ratio'            }      0.75       0.61       1.03 

% Create a stacked barplot of table Tindclean
barData = Tindclean{:, 2:end}; % Assuming the first column is categorical
barLabels = Tindclean{:, 1}; % Assuming the first column contains labels
figure;
bar(barData, 'stacked');
set(gca, 'XTickLabel', barLabels);
xlabel('Categories');
ylabel('Values');
title('Stacked Bar Plot of Tindclean');
legend(Tindclean.Properties.VariableNames(2:end), 'Location', 'bestoutside');

[Prashant joshi]

it looks like there is a scaling issue. There are some variables with different scales than others and i do not want to normalize them. can you please help?

% Create a stacked barplot of table Tindclean
barData = Tindclean{:, 2:end}; % Assuming the first column is categorical
barLabels = Tindclean{:, 1}; % Assuming the first column contains labels
figure;
bar(barData, 'stacked');
set(gca, 'XTickLabel', barLabels);
xlabel('Categories');
ylabel('Values');
title('Stacked Bar Plot of Tindclean');
legend(Tindclean.Properties.VariableNames(2:end), 'Location', 'bestoutside');

[Cris LaPierre]

If you don't want to normalize, then what strategy would you propose?

About the only option that I can think of, then, is to use logarithimic Y axis.

Tindclean = readtable('mydata.xls');
Tindclean.ValuationRatios = categorical(Tindclean.ValuationRatios);

% Create a stacked barplot of table Tindclean
barData = Tindclean{:, 2:end}; % Assuming the first column is categorical
barLabels = Tindclean{:, 1}; % Assuming the first column contains labels
figure;
bar(barLabels,barData, 'stacked');
xlabel('Categories');
ylabel('Values');
title('Stacked Bar Plot of Tindclean');
legend(Tindclean.Properties.VariableNames(2:end), 'Location', 'bestoutside');
yscale('log')

[Prashant joshi]

Thanks. That will work.

Can we set like yyaxis right yyaxis left may be for loop but I am not able to do that with the plot not sure if we can do it. Any insight is helpful.

[Cris LaPierre]

Possible without a for loop. However, the challenge now is how to indicate which bar corresponds to which axis.

Tindclean = readtable('mydata.xls');
Tindclean.ValuationRatios = categorical(Tindclean.ValuationRatios);

% Create a stacked barplot of table Tindclean
barData = Tindclean{:, 2:end}; % Assuming the first column is categorical
barLabels = Tindclean{:, 1}; % Assuming the first column contains labels
figure;
yyaxis left
colororder("gem")
ind = contains(string(barLabels),'Employee');
bar(barLabels(~ind),barData(~ind,:), 'stacked');
xlabel('Categories');
ylabel('Values');
yyaxis right
colororder("gem")
bar(barLabels(ind),barData(ind,:), 'stacked');

title('Stacked Bar Plot of Tindclean');
legend(Tindclean.Properties.VariableNames(2:end), 'Location', 'bestoutside');

●