formula sheet

# formula sheet - Statistical Inference for FE Mathematical...

This preview shows pages 1–2. Sign up to view the full content.

Statistical Inference for FE. Mathematical Formula Sheet. F -distribution F n 1 ,n 2 = χ 2 n 1 /n 1 χ 2 n 2 /n 2 , where χ 2 n 1 and χ 2 n 2 are two independent χ 2 distribution with d.f. n 1 and n 2 , respectively. F n 1 , = χ 2 n 1 /n 1 , where χ 2 n 1 is a χ 2 distribution with d.f. n 1 . Convergence in probability: X n converges to X in probability, if, for any ε > 0 , P ( | X n X | > ε ) 0 , as n →∞ . Law of Large Numbers and Central Limit Theorem. Suppose X 1 , X 2 , X 2 ,.... are i.i.d. random variables with mean μ = E [ X ] and variance σ 2 .Th en P n i =1 X i /n E [ X ] , a.s., P n i =1 X i σ n N (0 , 1) in distribution. Suppose that we try to f nd the root of f ( x )= a. The Newton algorithm is x m +1 = x m + a f ( x m ) f 0 ( x m ) . A standard one dimensional Brownian motion W ( t ) , t 0 ,start ingat 0 , is a continuous-time process with the properties that (1) W (0) = 0 , and the sample path of W ( t ) is continuous, almost surely; (2) W ( t 1 ) , W ( t 2 ) W ( t 1 ) , W ( t 3 ) W ( t 2 ) ,..., W ( t m ) W ( t m 1 ) are independent for every m 1 and 0 <t 1 <t 2 < ··· <t m < ;(3) W ( t ) W ( s ) is normally distributed with mean zero and variance t s . A Brownian motion W μ, σ ( t ) with drift μ and variance σ 2 is de f ned as W μ, σ ( t ):= μt + σ W t . Geometric Brownian motion: e at + σ W t , t 0 . Simpson’s paradox occurs when some overall conclusion concerning a set of objects fails to hold within each of a collection of subsets of those subjects. If the estimator ˆ θ has an asymptotic normal distribution ˆ θ N ( θ ,v ( ˆ θ )) , where v ( ˆ θ ) is the asymp- totic variance of ˆ θ ,then 95% con f dence interval for the unknown parameter θ is ( ˆ θ 1 . 96 q v ( ˆ θ ) , ˆ θ + 1 . 96 q v ( ˆ θ )) . An estimator ˆ θ of an unknown parameter θ is unbiased if E [ ˆ θ ]= θ . The bias of an estimator ˆ θ is de f ned to be E [ ˆ θ θ ] . Suppose we have two unbiased estimators ˆ θ 1 and ˆ θ 2 .W esay ˆ θ 1 is more e cient than ˆ θ 2 if Var ( ˆ θ 1 ) <Var ( ˆ θ 2 ) . The mean squared error of an estimator ˆ θ is MSE ( ˆ θ )= E [( ˆ θ θ ) 2 ] . Note that if θ is a scalar, then MSE ( ˆ θ )= Var ( ˆ θ )+( Bias ( ˆ θ ))

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 3

formula sheet - Statistical Inference for FE Mathematical...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online