Creating own header files for statistics

-

I was always interested in the subject of 'Probability & Statistics'. So I went ahead and created a header file to help me solve the questions present in the text book I was studying.

I've developed two header files:
  1. measures.h: It has all the functions to measure the various measures used in statistics like mean, median, variance, deviation etc.
  2. data.h: This contains the definitions of two data-types 'univariateData' and 'bivariateData'. These are helpful in processing data right on input and giving out details about the given data. They are immutable(can't be edited).
*Note: Don't include measures.h if data.h is already included because data.h has measures.h declared in it.

The cool part is that just 2 lines of code and user input are now enough to analyze any univariate data or bivariate data and display the info!


About the data: The given data is about the use of alternative fueled vehicles in the USA between the years 1992 and 1999. Here alternative fueled vehicles refer to vehicles which use ethanol, methanol, electricity, natural gas or LPG as the source of energy.

Here's the data(in tabular form):


YEAR NUMBER OF VEHICLES
1992 251,352
1993 314,848
1994 324,472
1995 333,049
1996 352,421
1997 367,526
1998 383,847
1999 407,542

Analyzing using the data.h file: 

Clearly, the given data is bivariate data as two values are being measured on an experimental unit. Here the experimental unit is the vehicle and the two variables being measured are it's year and number of such vehicles in that particular year.

So now, we can use the 'bivariateData' data-type provided by the data.h to store and analyze data. Furthermore, since the given data is bivariate data, we can also construct a regression line and predict the data! 

 Here's the program that I've written along with a snap of the execution:



Snap of execution:


*Note: In the regression equation(in the above snap), YEAR and NO_OF_VEHICLES were placed wrongly. It must be NO_OF_VEHICLES = (number) * YEAR + (number). I've made this change later.

So there you have it, the data has been stored and analyzed right on input! Similarly, this can be done
with univariate data. But we cannot predict data for univariate data as regression line cannot be constructed for it. We can only analyze.

So that's it for now. If you'd like to download the header files, here's the link to my google drive folder: https://drive.google.com/drive/folders/0B13xGrhLEVkjQm5aWV9IS3lJZ00?usp=sharing

By the way, you can directly use the functions such as mean(), median(), covariance() on any double[] array provided that you send the correct arguments. Have a look at the comments in header files to better understand.

Comments

Post a Comment

Popular posts from this blog

Beginner's guide to Solving the N-Queens problem using backtracking method

-
Image
This post serves as a good introduction to the backtracking method which is used widely in various types of problems to find a possible solution. Here we'll be seeing how to solve the classical N-Queens problem using backtracking method. Backtracking : The Wikipedia page  for Backtracking defines it as "a general algorithm in which, we try to incrementally build candidates to the solutions, and discard a candidate as soon as we get to know that the candidate cannot possibly be completed to a valid solution".  If that sounds complicated and confusing, don't worry! Let us understand it by running through a small example. Consider you are trapped in a haunted house with five rooms named A, B, C, D and E. Initially, you are in room A and you need to reach room E to exit the house. Below is an image of the rooms in the house. The arrows in the image indicate the directions in which you can move. For example, you can move to rooms B and C from A but, you
Read more

PvP Chain reaction game using Python

-
Image
At last, I've taken time and ventured a little deeper into the Python language! I used the  Tkinter   package to develop a rip-off of an android game called chain-reaction . It's a simple multiplayer board game in which the first person to occupy all the blocks on the board or of the opponent(s) wins. Try downloading and playing the game with your friends, it's FUN! Here's a video of the game I've developed(I've set the number of players to 4 and board dimensions are 6*8): Coming to the coding, I actually enjoyed the process of writing code much more in Python compared to any other previous languages that I've used. Maybe it is due to the fact that it's the first high-level language that I got my hands on. I'd never worked with Tkinter package before. It was quite easy to understand and I was able to build the game in roughly in four hours. I think that just a little effort is enough for anyone to understand how to use the  tkinter  packa
Read more

Developing Minesweeper using Java

-
Image
It's been a while since I've posted anything on my blog. The reason being that lately, I've been learning new languages like Python and Java. Well anyways, I've taken the time to develop Minesweeper game using Java applets. This game also got GUI which I've created using the components provided by Java. Here are a few pictures of the game. Here's the video of my game: There are two classes in my game: Board class : This class provides the backend logic of building the board for the minesweeper game. This class has also got methods using which user can play the game on command prompt itself. I've commented out the methods required for running the game on cmd ( the * beside a method listed below denotes that the method was commented ). If you're interested in playing on cmd, fiddle around with invoking methods in the main method(create one!) and see if you can run it :). The various methods in Board class are: Board() *:  Constr
Read more