My roller coaster prediction

**thex95** · 01-20-2012, 11:09 AM

This is just for fun and it does not represent the views, statements, or opinions of AR-R in any way.

So here is a linear regression chart using the historical data of the AR-R roller coaster sale.The vertical axis represents the % discount on AR-R products and the horizontal line represents the day of sale. The red line is the least squares regression line which is used to project (predict) numbers in the future. However the r² = .1187, meaning that all predictions to forecast future numbers are only 11.87% accurate.

So using the linear regression formula (y=-1.3929x + 29.143) it can be predicted that on Monday (day 8 of the roller coaster) the sale will be for 17.9998% and on Tuesday (day 9 of the roller coaster) for 16.6069%. Once again these predictions are only 11.87% accurate due to the high fluctuation in sale percentages during this sale.

Once again- This is just for fun and it does not represent the views, statements, or opinions of AR-R in any way.

**JohnnyVegas** · 01-20-2012, 01:02 PM

Fun dorky post. And I mean "dorky" in a good way.

Unfortunately, they can change their plans just to spite you. Fortunately, you can claim that is exactly what they did if you predictions do not turn out to be accurate.

**tballz** · 01-20-2012, 01:04 PM

Originally Posted by scotts

This is just for fun and it does not represent the views, statements, or opinions of AR-R in any way.

So here is a linear regression chart using the historical data of the AR-R roller coaster sale.The vertical axis represents the % discount on AR-R products and the horizontal line represents the day of sale. The red line is the least squares regression line which is used to project (predict) numbers in the future. However the r² = .1187, meaning that all predictions to forecast future numbers are only 11.87% accurate.

So using the linear regression formula (y=-1.3929x + 29.143) it can be predicted that on Monday (day 8 of the roller coaster) the sale will be for 17.9998% and on Tuesday (day 9 of the roller coaster) for 16.6069%. Once again these predictions are only 11.87% accurate due to the high fluctuation in sale percentages during this sale.

Once again- This is just for fun and it does not represent the views, statements, or opinions of AR-R in any way.

haha....that's pretty good analysis. So you're saying buy now?

**RUI-Products** · 01-20-2012, 01:51 PM

Originally Posted by scotts

This is just for fun and it does not represent the views, statements, or opinions of AR-R in any way.

So here is a linear regression chart using the historical data of the AR-R roller coaster sale.The vertical axis represents the % discount on AR-R products and the horizontal line represents the day of sale. The red line is the least squares regression line which is used to project (predict) numbers in the future. However the r² = .1187, meaning that all predictions to forecast future numbers are only 11.87% accurate.

So using the linear regression formula (y=-1.3929x + 29.143) it can be predicted that on Monday (day 8 of the roller coaster) the sale will be for 17.9998% and on Tuesday (day 9 of the roller coaster) for 16.6069%. Once again these predictions are only 11.87% accurate due to the high fluctuation in sale percentages during this sale.

Once again- This is just for fun and it does not represent the views, statements, or opinions of AR-R in any way.

scotts, love the regression model. And good job on it, btw. Given that the conditional expectation of probability gives you predictions that are only 11.87% accurate, I'd say it gives me quite a bit of leeway. Here is what I can promise....that the high fluctuations in discount between each day will continue to occur, causing even more variability in the regression model. However, as the rollercoaster continues forward, each day will give you additional independent variable data..who knows, you maybe able to increase the r² to MUCH higher by the end of this thing.

btw, check your PM's. I'm sending you a personalized promocode for creativity.

AR-R