1. |
|
|
2. |
|
|
3. |
|
|
4. |
|
|
5. |
- Brännäs, Kurt, 1949-
(författare)
-
On estimation in econometric systems in the presence of time-varying parameters
- 1980
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Economic systems are often subject to structural variability. For the achievement of correct structural specification in econometric modelling it is then important to allow for parameters that are time-varying, and to apply estimation techniques suitably designed for inference in such models. One realistic model assumption for such parameter variability is the Markovian model, and Kaiman filtering is then assumed to be a convenient estimator. In the thesis several aspects of using Kaiman filtering approaches to estimation in that framework are considered. The application of the Kaiman filter to estimation in econometric models is straightforward if a set of basic assumptions are satisfied, and if necessary initial specifications can be accurately made. Typically, however, these requirements can generally not be perfectly met. It is therefore of great importance to know the consequences of deviations from the basic assumptions and correct initial specifications for inference, in particular for the small sample situations typical in econometrics. If the consequences are severe it is essential to develop techniques to cope with such aspects.For estimation in interdependent systems a two stage Kaiman filter is proposed and evaluated, theoretically, as well as by a small sample Monte Carlo study, and empirically. The estimator is approximative, but with promising small sample properties. Only if the transition matrix of the parameter model and an initial parameter vector are misspecified, the performance deteriorates. Furthermore, the approach provides useful information about structural properties, and forms a basis for good short term forecasting.In a reduced form fraaework most of the basic assumptions of the traditional Kaiman filter are relaxed, and the implications are studied. The case of stochastic regressors is, under reasonable additional assumptions, shown to result in an estimator structurally similar to that due to the basic assumptions. The robustness properties are such that in particular the transition matrix and the initial parameter vector should be carefully estimated. An estimator for the joint estimation of the transition matrix, the parameter vector and the model residual variance is suggested and utilized to study the consequences of a misspecified parameter model. By estimating th transitions the parameter estimates are seen to be robust in this respect.
|
|
6. |
|
|
7. |
- Brännäs, Kurt
(författare)
-
Prediction in a duration model
- 1986
-
Ingår i: Journal of Forecasting. - : Wiley. - 0277-6693 .- 1099-131X. ; 5:2, s. 97-103
-
Tidskriftsartikel (refereegranskat)
|
|
8. |
- Malmgren, Helge, 1945
(författare)
-
On the nature of reinforcement.
- 1985
-
Ingår i: Göteborg Psychological Reports. - 0301-0996. ; 16:3
-
Tidskriftsartikel (övrigt vetenskapligt/konstnärligt)abstract
- In this paper I model learning through negative reinforcement as a kind of random walk and report some simulations of the model.
|
|
9. |
|
|
10. |
- Nowicki, Krzysztof
(författare)
-
Asymptotic normality of graph statistics
- 1989
-
Ingår i: Journal of Statistical Planning and Inference. - 1873-1171. ; 21:2, s. 209-222
-
Tidskriftsartikel (refereegranskat)abstract
- Various types of graph statistics for Bernoulli graphs are represented as numerators of incomplete U-statistics. Asymptotic normality of these statistics is proved for Bernoulli graphs in which the edge probability is constant. In addition it is shown that subgraph counts asymptotically are linear functions of the number of edges in the graph.
|
|
11. |
|
|
12. |
|
|
13. |
- Nowicki, Krzysztof
(författare)
-
Asymptotic Poisson Distributions with Applications to Statistical Analysis of Graphs
- 1988
-
Ingår i: Advances in Applied Probability. - 0001-8678. ; 20:2, s. 315-330
-
Tidskriftsartikel (refereegranskat)abstract
- Various types of graph statistics for graphs and digraphs are presented as numerators of incomplete U-statistics, with symmetric and asymmetric kernels, respectively. Thus, asymptotic Poisson limits of these statistics are provided by using limit theorems for the sums of dissociated random variables. Several applications to statistical analysis of graphs are given
|
|
14. |
- Björck, Svante, et al.
(författare)
-
A magnetostratigraphic comparison between 14C years and varve years during the late Weichselian, indicating significant differences between the time-scales
- 1987
-
Ingår i: Journal of Quaternary Science. - : Wiley. - 1099-1417 .- 0267-8179. ; 2:2, s. 133-140
-
Tidskriftsartikel (refereegranskat)abstract
- A 14C-dated magnetostratigraphy of absolute declination and inclination between 12500 and 10000 14C yr BP was recently developed for southern Sweden. Recently also the Swedish geochronological time-scale, based on c. 11 500 annually deposited clay-varves, was connected with the present. It should therefore be possible to compare the two chronologies with a reliable magnetostratigraphic record in an appropriate clay-varve section. We have found such a site within the Middle Swedish end-moraine zone. Statistical correlations between the two independently dated time-scales suggest that at 10500–10200 14Cy r BP the varve chronology exceeds the 14C chronology by the order of 500-600 varve yr. Other correlations indicate that the difference between the two chronologies was less at 11000 14C yr BP, and further correlations between the time-scales at 12000 14C yr BP suggest that the difference between the chronologies increased steadily from 12000 to 10000 14C yr BP. If these correlations are correct they imply that the 14C production rate increased steadily during the Late Weichselian.
|
|
15. |
- Ekblom, Håkan
(författare)
-
A new algorithm for the huber estimator in linear models
- 1988
-
Ingår i: BIT Numerical Mathematics. - 0006-3835 .- 1572-9125. ; 28:1, s. 123-132
-
Tidskriftsartikel (refereegranskat)abstract
- This paper considers algorithms for solving the linear robust regression problem by minimizing the Huber function. In the computational methods for this problem used so far, the scale estimate is adjusted separately. The new algorithm, based on Newton's method, treats both the scale and the location parameters as independent variables. The special form of the Hessian allows for an efficient updating scheme.
|
|
16. |
- Ekblom, Håkan, et al.
(författare)
-
Algorithms for non linear huber estimation
- 1989
-
Ingår i: BIT Numerical Mathematics. - 0006-3835 .- 1572-9125. ; 29:1, s. 60-76
-
Tidskriftsartikel (refereegranskat)abstract
- The Huber criterion for data fitting is a combination of thel 1 and thel 2 criteria which is robust in the sense that the influence of wild data points can be reduced. We present a trust region and a Marquardt algorithm for Huber estimation in the case where the functions used in the fit are non-linear. It is demonstrated that the algorithms converge under the usual conditions.
|
|
17. |
|
|
18. |
|
|
19. |
|
|
20. |
|
|
21. |
- Lindgren, Georg
(författare)
-
Point processes of exits by bivariate Gaussian processes and extremal theory for the chi^2-process and its concomitants
- 1980
-
Ingår i: Journal of Multivariate Analysis. - 0047-259X. ; 10:2, s. 181-206
-
Tidskriftsartikel (refereegranskat)abstract
- Let ζ(t), η(t) be continuously differentiable Gaussian processes with mean zero, unit variance, and common covariance function r(t), and such that ζ(t) and η(t) are independent for all t, and consider the movements of a particle with time-varying coordinates (ζ(t), η(t)). The time and location of the exists of the particle across a circle with radius u defines a point process in R3 with its points located on the cylinder {(t, u cos θ, u sin θ); t ≥ 0, 0 ≤ θ < 2π}. It is shown that if r(t) log t → 0 as t → ∞, the time and space-normalized point process of exits converges in distribution to a Poisson process on the unit cylinder. As a consequence one obtains the asymptotic distribution of the maximum of a χ2-process, χ2(t) = ζ2(t) + η2(t), P{sup0≤t≤T χ2(t) ≤ u2} → e−τ if T(−r″(0)/2π)1/2u × exp(−u2/2) → τ as T, u → ∞. Furthermore, it is shown that the points in R3 generated by the local ε-maxima of χ2(t) converges to a Poisson process in R3 with intensity measure (in cylindrical polar coordinates) (2πr2)−1 dt dθ dr. As a consequence one obtains the asymptotic extremal distribution for any function g(ζ(t), η(t)) which is “almost quadratic” in the sense that Image Full-size image has a limit g*(θ) as r → ∞. Then P{sup0≤t≤T g(ζ(t), η(t)) ≤ u2} → exp(−(τ/2π) ∫ θ = 02π e−g*(θ) dθ) if T(−r″(0)/2π)1/2u exp(−u2/2) → τ as T, u → ∞.
|
|
22. |
|
|
23. |
|
|
24. |
|
|
25. |
|
|