Curve fitting, as the name implies, is about fitting a curve that best explains the 'relationship' between two 'quantitative' variables. With experimental data, the relationship between the response variable (Y) and quantitative' treatment factor (X) can be derived using a number of freely available software such as R. I do not use Origin, so difficult for me to say how that works to fit a curve. May be talk to someone in your Stats department to help you. 

Dear Pallavi Patil,

Curve fitting is the process of constructing a curve, or mathematical functions, which possess closest proximity to the series of data. By the curve fitting we can mathematically construct the functional relationship between the observed fact and parameter values, etc. It is highly effective in mathematical modelling some natural processes.

Many of the recent mathematical software do the job efficiently. But I like to use the GNUPLOT for this purpose. The fitting command for gnuplot is

==========================================================

f(x) =m*sin(n*x) # define the function to be fit

n=0.5; m=2.2 # initial guess for a1 and b1

fit f(x) 'filename.txt' using 1:2 via n, m  #fits the datafile to the function

plot "filename.txt" u 1:2 w lp, f(x) #plot the datafile along with the function.

============================================================

By this you may fit the data to the mathematical functions. It optimizes the parameters associated with the defined function and plots it along with the data.

Best of luck.

Debabrata Biswas.

When you have experimental data and you are willing to find the relation between the X and Y parameters, then you use one of the methods of the least squares methods to find the polynomial the has the least distance from the data; the regression line is the case of polynomial of degree 1, i.e. a line. Polynomials of degrees 2 and 3 give better result and also the exponential and logarithmic approximations might be more proper depending on the data behavior. See the link:

http://en.wikipedia.org/wiki/Least_squares_%28function_approximation%29