gtcam

03-25-2008, 12:05 AM

I'm looking to build a model in excel which will need some statistical formulas, but i'm not confident i can do this without some help or clarification from some experts. I have a technical background (Engineering) but its been alot of years since those statistics classes in college and i'm a bit rusty.

Anyway, Here is what i'm doing:

At a given location, market based electricity prices change on a daily basis based on a number of factors. The primary factor is demand (usage). Various factors affect the demand for electricity, primarily temperature, time of day, day of week, and how many powerplants are in outage at any given time. These power plant outages are both planned events, and random events. the bottom line is that its a complicated array of variables that cause the energy market to move as they do, and anyone that could develop the equation to predict this could obviously make alot of money in the markets. What i want to do is accept the fact that its complicated, but analyze a large historical sample to gain confidence of where prices could potentially be at any given time.

For example, at location A, i have 3 years of historical prices. These prices are hourly, so at location A, there are 24 different prices each day (hour 1-24). I want to do this analysis for each of the 24 hours independently, but using hour 1 as an example, I want to look at hour 1 for each day of the 3 years of sample data to come up with statistics to show the range of possibilities as to where prices could be with various levels of confidence. Like a price level that i'm 95% confident that prices will be below that level. Or another price level where i'm 95% sure that prices will be ABOVE the given price.

can anyone get me going with this? i'm very proficient in excel, and if i can get the relevent concepts and formulas, i'm sure i can build the model in excel.

Thanks!

Anyway, Here is what i'm doing:

At a given location, market based electricity prices change on a daily basis based on a number of factors. The primary factor is demand (usage). Various factors affect the demand for electricity, primarily temperature, time of day, day of week, and how many powerplants are in outage at any given time. These power plant outages are both planned events, and random events. the bottom line is that its a complicated array of variables that cause the energy market to move as they do, and anyone that could develop the equation to predict this could obviously make alot of money in the markets. What i want to do is accept the fact that its complicated, but analyze a large historical sample to gain confidence of where prices could potentially be at any given time.

For example, at location A, i have 3 years of historical prices. These prices are hourly, so at location A, there are 24 different prices each day (hour 1-24). I want to do this analysis for each of the 24 hours independently, but using hour 1 as an example, I want to look at hour 1 for each day of the 3 years of sample data to come up with statistics to show the range of possibilities as to where prices could be with various levels of confidence. Like a price level that i'm 95% confident that prices will be below that level. Or another price level where i'm 95% sure that prices will be ABOVE the given price.

can anyone get me going with this? i'm very proficient in excel, and if i can get the relevent concepts and formulas, i'm sure i can build the model in excel.

Thanks!