Friday 19 July 2013

SciStatCalc version 1.2 released

Version 1.2 of SciStatCalc now available on iTunes, with the file import bug fixed.

Saturday 13 July 2013

SciStatCalc version 1.1 released

An update to SciStatCalc (version 1.1) is now available on iTunes. As part of the updates, it is now possible to import a two-column CSV file as an email attachment, into the pink statistical tests textviews.

Update: Have just discovered a bug that restricts the first column data of the CSV file to positive integers, and have fixed this for the next version (1.2), which I have submitted and is currently waiting for review. Nevertheless, I will add a few screenshots below to illustrate the file import facility.

As an example, I have emailed a two column, 16 row CSV file, as below.

Importing email attachment CSV file into iOS App

Opening the attachment results in:-

CSV file opened
Clicking on the top right button results in a list of apps capable of importing the attachment, starting with our app.


Opening the attachment in SciStatCalc (by pressing on the SciStatCalc app icon), then orienting the iPhone in landscape mode, and finally swiping to the desired statistical test display results in the screenshot below.


Pressing the calculate button, will result in the following screenshot, where the results of the unpaired Student's t-test are displayed below the textviews:-




In tandem, I have updated the pdf (document format, no pun intended!) file with a list of probability distributions (19 in all, including two recent additions for version 1.1 - Pareto and Hypergeometric) and their CDF. The pdf file (embedded directly below) has a third page, with some useful notes including the expression for the CDF of the Hypergeometric function, as well as the definitions of the more specialised functions that form the CDF of many of the listed distributions.


Logistic Regression Calculator and ROC Curve Plotter

This blog post implements a Logistic Regression calculator for a binary output. Consider a binary outcome response variable \(Y\...