How can I change 'y axis' into 'Probability', not the 'number' when I use 'hist(variable)' ?

Figure 1 is what I get from hist(variable);

Figure 2 is what I want to get;

Figure 3 is what I want to to get as well.

[Walter Roberson]

Is it truly necessary to use hist() instead of histogram() ? histogram() has normalization options built in.

[wenhao yang]

Yes I've already found it. Then How can I get something like Figure 3 ?

[Walter Roberson]

For figure 3 you either need more bins or you need to fit a model.

[wenhao yang]

So can you offer me some example on how to write the command to get something like Figure 3 ?

[Walter Roberson]

https://www.mathworks.com/help/stats/histfit.html

h = histfit(___) returns a vector of handles h, where h(1) is the handle to the histogram and h(2) is the handle to the density curve.

So you could set the 'Normalization' option of h(1) to change how the axes got labeled. It is possible that you will need to rescale the curve on h(2) as well.

Answer 1

[KALYAN ACHARJYA]

Is there any sense? Let's suppose I have the following array

> data
data =
     4     3     5     3     5     5     3     4     3     5     2     5     2     1     4     5     5     2     5     3

Now the bins and counts as follows

bins:    1     2     3     4     5
counts:  1     3     5     3     8

If you plot the hist of the data, no issue

Once you wish to reverse bar plot, then issue arrise with random data, which is not unique (It suppose to be unique for variable data). However you can get the line plot in this case

[counts,bins]=hist(data); 
plot(counts,bins);

Counts data in the variable axes have two issue.

• Non unique data
• No linear order (Incresing or Decreasing)

Although you can do other ways to plot the histogram appearance, like barh etc.

Hope it Helps! :)