1
Understanding ChiSquared
• Consider 2 situations:
– Comparing sampled Data directly with the parent
distribution from which it came
– Comparing sampled Data with our best estimate of
that parent distribution
2
1
2
)
;
(
∑
=
⎥
⎦
⎤
⎢
⎣
⎡
−
=
N
i
i
i
i
i
a
x
f
y
σ
χ
2
1
2
2
)
;
(
1
reduced
∑
=
⎥
⎦
⎤
⎢
⎣
⎡
−
−
=
=
N
i
i
i
i
i
a
x
f
y
M
N
DOF
Performing a LeastSquares Fit
and Understanding the Result
A complete example
Δ
V = 0.05 V
for all
measurements
()
2
1
2
∑
=
⎥
⎦
⎤
⎢
⎣
⎡
Δ
+
−
=
N
i
i
i
i
y
b
ax
y
SUMMARY: Best Fit line with
χ
2
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document2
Compute ChiSquare for fit
()
2
1
2
∑
=
⎥
⎦
⎤
⎢
⎣
⎡
Δ
+
−
=
N
i
i
i
i
y
b
ax
y
χ
= 1.95 (for example data on previous slide)
DF = Degrees of Freedom
P = Probability of exceeding the tabulated value of
Chisquare (for a given DF)
How do we interpret this value of
ChiSquare?
• We interpret this the following way:
If we were
to repeat the same experiment many times, we
would expect approximately 96% of those
This is the end of the preview. Sign up
to
access the rest of the document.
 Fall '07
 Dobbs
 Approximation, 1 Hour, physicist, 4%

Click to edit the document details