Scribd
Upload
74 views10 pages

The Gaussian Distribution

The document discusses the Gaussian distribution and its properties. The Gaussian distribution results from variation caused by many independent factors and follows a symmetrical bell curve. It is characterized by its mean and standard deviation, with 68% of values falling within 1 standard deviation of the mean and 95% within 1.96 standard deviations. While large populations often approximate a Gaussian distribution, smaller samples require calculating a prediction interval to account for the population potentially deviating from perfect Gaussian form.

Uploaded by

abrar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
74 views10 pages

The Gaussian Distribution

The document discusses the Gaussian distribution and its properties. The Gaussian distribution results from variation caused by many independent factors and follows a symmetrical bell curve. It is characterized by its mean and standard deviation, with 68% of values falling within 1 standard deviation of the mean and 95% within 1.96 standard deviations. While large populations often approximate a Gaussian distribution, smaller samples require calculating a prediction interval to account for the population potentially deviating from perfect Gaussian form.

Uploaded by

abrar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 10

The

Gaussian
Distribution
Probability distribution

Histograms are used to plot the distribution of values in a sample. It can be bell shaped.

But if the population is infinitely large, a probability distribution is seen.

Because of large population, the Y-axis in probability distribution is known as probability density as it
cannot show the number of observations in each group
The Gaussian Distribution

 The symmetrical bell-shaped distribution is called a Gaussian distribution

 The variables will approximate a Gaussian distribution when the variation is caused by many
independent factors
 When the variation is largely due to one factor (such as variation in a single gene), you expect to see
bimodal or skewed distributions rather than Gaussian distributions.
The Gaussian Distribution

The mean is the centre of the Gaussian distribution and standard deviation is the spread of distribution

The area under the curve within 1 SD represents 68% of the total observation and the area under the
entire curve represents the entire population

95% of the values in a Gaussian distribution lie within 1.96 SD of the mean.

Areas under a portion of the Gaussian distribution are tabulated as the "z' ' distribution, where z is the
number of standard deviations away from the mean.

For each value of z, the table shows four columns of probabilities.


The Gaussian Distribution
USING THE GAUSSIAN DISTRIBUTION TO MAKE
INFERENCES ABOUT THE POPULATION

For any value Y, you can calculate z using the Equation

There are two problems with these calculations. One problem is that you don’t know the mean and SD of
the entire population. The second problem is that the population may not really be Gaussian

By calculating the z-value we can make other calculations for the percentage of area
THE PREDICTION INTERVAL

The prediction interval is only valid when the population is distributed according to a
Gaussian distribution
THE PREDICTION INTERVAL

To calculate a prediction interval, you need to go K SDs from the mean, where K is obtained from
Table
NORMAL LIMITS

The term normal limits can be defined in many ways. Using the Gaussian distribution to make rules
about "normal“ values is not always useful.

The deviations from Gaussian are likely to be most apparent in the tails of the distribution. There is
also no reason to think that the population distribution is symmetrical around the mean.

There are complexities of "normal limits" and problems in defining normal limits as being the

mean plus or minus 2 SDs


Comments


  With large enough sample size, the distribution of a population can be predicted which is known as
the probability distribution
 The Gaussian distribution is basically a normal distribution which is dependent on many independent
factors for its variation

 The Gaussian distribution follows 68-95-99.7 rule

 Any distribution can be converted to Gaussian distribution using equation

 In case of a small sample having Gaussian distribution, prediction interval needs to be used which
means to use more than 2SD for including 95% of the observations

You might also like