Random Values

Until now we have not used any real data for creating plots. Therefore we will start with some random values. In order to create them, we use the built-in class TRandom2 which shows the best performance according to the official ROOT documentation. We will get access to this by first creating a new file called randomize.C and then creating an object called r in the following way:

 TRandom2 *r = new TRandom2();

Now we can create a uniform distribution by using the function Rndm():

 double rval = r->Rndm();

This function returns a value between 0 and 1 which will then be stored in the double variable rval. The full program then looks like that

 void randomize()
{
    TRandom2 *r = new TRandom2();
    double rval = r->Rndm();
    cout << rval << endl;
}

Now we observe, that whenever we run the program, the value 0.186915 is given out in the terminal. The reason for this can be found in the fact that in the standard configuration, TRandom2 uses a fixed seed. If we change the first for instance to

 TRandom2 *r = new TRandom2(2);

then the output value looks different but is still constant whenever we start the program. If we want to avoid this, we can set a random seed with the following code line:

 TRandom2 *r = new TRandom2(0);

Instead of using the function Rndm(), we can also use Gaus() which creates random numbers from a Gaussian distribution. We can additionally insert two parameters: the first one is the mean value $\mu$ and the second one is the standard deviation $\sigma$. Now we want to create a loop with 1000 iterations and in each step, we will our random value into a histogram with 100 bins and the limits -5 and 5. The full script looks as follows:

void randomize()
{
    TRandom2 *r = new TRandom2(0);
    TH1F *hist = new TH1F("hist", "Histogram", 100, -5, 5);
    for(int i = 0; i < 1000; i++)
    {
        double rval = r->Gaus(0., 0.5);
        hist->Fill(rval);
    }
    TCanvas *c1 = new TCanvas();
    hist->Draw();
}

The result is a nice-looking Gaussian centered around 0.