From d7219d3a90101417c9e9ec8f8789e1150741edc1 Mon Sep 17 00:00:00 2001
From: jeffchiu22 <37814371+jeffchiu22@users.noreply.github.com>
Date: Mon, 16 Jul 2018 23:41:44 +0800
Subject: [PATCH] updated

go warriors
---
 Business Analytics/README.txt | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/Business Analytics/README.txt b/Business Analytics/README.txt
index e69de29..7742473 100644
--- a/Business Analytics/README.txt	
+++ b/Business Analytics/README.txt	
@@ -0,0 +1,21 @@
+For the business analytics file, we have used data about viewership per date and viewership per team to create our predictions.
+
+Using Date
+
+On the Date VPG sheet, we have organized it such that the numbers displayed are the average numbers of viewers per game per day. We have
+then created graphs, one for each season, which creates a cubic trendline as well as a variance value.
+
+Using Team
+
+On the Team VPG sheet, we have organized it such that the numbers displayed are the average numbers of viewers per game per team. We have
+then created graphs, one for each season, which creates a cubic trendline as well a variance value.
+
+Combining Data
+
+We will collect 3 data points and 3 variance values to predict the viewership for the game. From the two teams that are playing against
+each other, we will use the appropriate team trendline from the Team VPG sheet to get an average viewership value from each team. From the
+date of the game, we will use the appropriate date trendline from the Date VPG sheet to get an average viewership value for that specific
+date. The average viewership values of the teams and the average viewership value of the date will make up the 3 data points, and the 2
+variance values of the teams along with the variance value of the date will make up the variance 3 variance values. Then, to get the
+final predicted viewership value, we will multiple each data point by the fraction of the inverse of its corresponding variance value over
+the sum of all inversed variance values.