Expert: Sherman D. - 2/26/2004

Question
Thanks, i probably should have emphasized the statistics part of my question. Excel can calculate the slope (m) and any intercept for me, but how can i test if this is significantly different from a zero slope? I could send you the data but i need to repeat this test for 100 data sets. Each data set simply has about 90 values - one for each year. So i guess i'm looking for advice on the appropriate statistical test to use.

Thanks again.

-------------------------
Followup To
Question -
Hello, i hope you can help me.

I need to analyse a data set (I have approx 90 years of data and a single value for each year). I need to determine if a trend is apparent - i suppose i need to ask the question "do the data show a trend/slope that is significantly different from zero". But i'm not sure how to test for this.

Thanks
D.

Answer -
since i don't have the data, this is all i can tell you

(x1,x2) and (y1,y2)

m(slope) = (y2 - y1)/(x2 - x1)

if you would like the formula, then you have

y = mx + b

y = ((y2 - y1)/(x2 - x1))x + b

which would become

y1 = (((y2 - y1)/(x2 - x1))(x1)) + b

to find b, you would do this

b = y1 - (((y2 - y1)/(x2 - x1))(x1))

so now you have

y = ((y2 - y1)/(x2 - x1))x + (y1 - (((y2 - y1)/(x2 - x1))(x1)))

but since you only want to know if the have the same slope, then all you need to do is take the formula

m = ((y2 - y1)/(x2 - x1))

don't understand, then you will have to give me some of the data.

to find out if each data works, you will have to use the y = mx + b formula.

Answer
y = mx + b
y = 0(x) + b
y = 0 + b
y = b

so if your y-intercept is equal to your y value.

or to find that mathematically, lets say you have

(x1,y)(x2,y), where as the y's are the same

m = (y - y)/(x2 - x1)
m = 0/(x2 - x1)

Not much more i can do.
m = 0

If you were to graph it, you would get a horizontal line that passes through the y-axis.