Line of Best Fit / Linear Regression

Question

I am trying ot get a line of best fit for a set of data in oracle
Lets say i have the following dataset.
Since there are a lot of 0 in the Terms I would like to ignore those Terms
Only use the Terms which have values to get a line of best fit.
Any help will be appreciated.
Thank you.
       TERMS            Count
TERM_0801     78
TERM_0802     58
TERM_0803     0
TERM_0804     82
TERM_0805     0
TERM_0806     0
TERM_0807     32
TERM_0808     0
TERM_0901     92
TERM_0902     0
TERM_0903     0
TERM_0904     56
TERM_0905     0
TERM_0906     0
TERM_0907     0
TERM_0908     0
TERM_1001     85
TERM_1002     0
TERM_1003     0
TERM_1004     67
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bi
Edited by: Chloe_19 on Jul 24, 2012 11:14 PM

Answer 1

Argh, i wonder about the values i got, different from what excel says ;-), till you came up with your solution.
They changed the order of the paramters from that what i consider as common:
"expr1 is interpreted as a value of the dependent variable (a y value), and expr2 is interpreted as a value of the independent variable (an x value)."
Do they also have left-hand traffic in the new world?
However, didnt OP wanted to have the zeros eliminated before the calculus?
with data as
select 'TERM_0801' t, 78 y from dual union all
select 'TERM_0802', 58 from dual union all
select 'TERM_0803', 0 from dual union all
select 'TERM_0804', 82 from dual union all
select 'TERM_0805', 0 from dual union all
select 'TERM_0806', 0 from dual union all
select 'TERM_0807', 32 from dual union all
select 'TERM_0808', 0 from dual union all
select 'TERM_0901', 92 from dual union all
select 'TERM_0902', 0 from dual union all
select 'TERM_0903', 0 from dual union all
select 'TERM_0904', 56 from dual union all
select 'TERM_0905', 0 from dual union all
select 'TERM_0906', 0 from dual union all
select 'TERM_0907', 0 from dual union all
select 'TERM_0908', 0 from dual union all
select 'TERM_1001', 85 from dual union all
select 'TERM_1002', 0 from dual union all
select 'TERM_1003', 0 from dual union all
select 'TERM_1004', 67 from dual
select
round(regr_slope(y,x), 4) slope
,round(regr_intercept (y, x) , 4) intercept
,round(regr_r2(y,x), 4) best_fit
from (
select
  row_number() over ( order by t ) x, y
from data
where
y!=0
SLOPE INTERCEPT BEST_FIT
.4762 66.6071 .0035 regards
Edited by: chris227 on 25.07.2012 04:58
Edited by: chris227 on 25.07.2012 05:10
best fit added