Site Visited 498534 times Page Visited 30 times You are in : Etantonio/EN/Universita/2anno/TeoriaFenomeniAleatori/     

Mathematical statistics

1) Esteem for intervals :

Draft of a having esteem the objective to characterize between all i values of c1 and c2 for which {c1 < X < 2c } = g, that one of minimal length.

 

2) Verification of the hypotheses statistics:

Branch of the theory of the decision based on the choice between 2 functions of distributions of probability is one.

H0 said base hypothesis

H1 said hypothesis alternative

 

3) Spazio of the events :

It is with of the possible events between which to choose, is indicated with M.

 

4) Spazio of marks them :

It is with of all the possible shapes of wave sended to the observer, is indicated with S.

 

5) Spazio of the observations :

It is with of all the possible ones marks them receipts, is indicated with Z.

 

6) Spazio of the decisions :

It is with of all the possible ones turns out you of the decision process, is indicated with D.

 

7) Rule of decision :

Draft of a determinist rule that associates to every observation one decision.

 

8) Error of 1° type in one binary decision after single observation :

An error of 1° type is had if the hypothesis H0 also comes rejected being true.

 

9) Error of 2° type in one binary decision after single observation :

An error of 2° type is had if the correct hypothesis is H1 but it turns out you of the experiment do not support the refusal of H0.

 

10) Criterion of the maximum verosimiglianza :

The decision rule chooses the event that has greater probability to have caused the observation.

 

11) Relationship of verosimiglianza :

 

12) Criterion of Neyman - Person :

It characterizes a rule of decision that diminishes b , having fixed to .

 

13) Difference between parameter esteem and verification of the hypotheses :

The parameter esteem concurs through the observations on a single model to estimate of or more parameters while the verification of the hypotheses puts to comparison 2 models, that one of the null hypothesis and that one of the hypothesis alternative.

Accidental numbers

14) operating Definition of accidental number :

Sequence of accidental numbers is one turning out from a physical experiment

 

15) conceptual Definition of accidental number :

Sequence of variable aleatory is one i.i.d.

 

16) Modality for the generation of accidental numbers :

to)    through tables

b)    through congruenziali algorithms

 

17) Describe a generator of accidental numbers ricorsivo :

Part from a value begins them and si it obtains the successive value in function of the previous value.

 

18) Describe a generator of accidental numbers congruenziale :

The present value is based on the congruenza concept therefore them is legacy to the rest of the division for m of the previous value.

 

19) Method in order to generate accidental numbers with one given distribution through a calculating:

(u) is necessary to calculate the values of the Ffunction-1 for u = ui , being {u} one sequence of accidental numbers uniform.

 

20) Describe the method Mount Carl :

It is based on an accidental sampling and used as an example for the calculation of integrals, in practical N is carried out times an aleatory experiment and then it is estimated the average of all turns out obtained to you.

Theory of the esteem

21) Esteem of point :

Draft of one function of the observation x it è = g(x)

 

22) Appraiser :

Draft of one function of carrier X constituted from all the observations x it è = g(X)

 

23) Error of esteem :

It is the difference between the appraiser and the incognito parameter q cioè and = - q

 

24) Polarization of the appraiser:

Draft of the valor medium of the error of stima b() = E(- q)

 

25) Variance of one appraiser :

It is the expected value of the square of the error of esteem V() = E[ (- E())2 ]

 

26) consisting Appraiser:

An appraiser is consisting if the esteem error stretches to 0 for N that stretches to infinite.

 

27) optimal Appraiser :

He is the appraiser who diminishes the quadratic error medium e = E[ (- q)2 ]

 

28) Describe the esteem for intervals:

They come defined a brace of functions of the observation1 = g 1 (X) e 2 = g 2 (X) which define a intervallo to the inside of which the parameter q is contained with one probability g.

 

29) Objective of the esteem for intervals :

To diminish the length of the interval |1 - 2 | .