Ola Hössjer


Bild på Ola
Welcome!

My name is Ola Hössjer and I'm Professor of
Mathematical Statistics at Stockholm University.

I have two daughters, Evelina and Linnea, and live in Sollentuna
north of Stockholm.

Some Recent Teaching

Fall 2005 : Stochastic methods of population genetics  (in English)
Spring 2006 and Fall 2007: Probability theory III (in Swedish 2006, English 2007)
Fall 2006: Stochastic processes and simulation II (in English)
Spring 2007: Stochastic processes and simulation I (in Swedish)
Fall 2007: Graduate Course in Probability Theory (in English)
Spring 2008: Population genetics and gene mapping (in English)
Fall 2008: Stochastic Processes III  (in English)

Research

My research has focused on various topics in
Publications
Grünewald, M. and Hössjer, O. (2011). A general statistical framework for multistage designs. To appear in Scandinavian Journal of Statistics. R-files (written by Maria Grünewald) available here: Efficiency_three-stage.R, Efficiency_two-stage_linear_regression.R and Efficiency_two-stage_logistic_regression.R.

Björkwall, S., Hössjer, O., Ohlsson, E. and Verrall, R. (2011). A generalized linear model with smoothing effects for claims reserving. Insurance, Mathematics and Economics, 49, 27-37.

Grünewald, M., Hössjer, O. and Humphreys, K. (2010). A Stochastic EM type algorithm for estimation in data with ascertainment on continuous outcomes. Report 2008:5, Mathematical Statistics, Stockholm University. International Journal of Biostatistics 6(1), Article 23.

Grünewald, M. and Hössjer, O. (2010). Efficient ascertainment schemes for maximum likelihood estimation. Journal of Statistical Planning and Inference 140(7), 2078-2088.

Björkwall, S., Hössjer, O. and Ohlsson, E. (2010). Bootstrapping the separation method in claims reserving. ASTIN Bulletin 40(2), 845-869.

Hartman, L., Hössjer, O and Humphreys, K. (2009). Ancestral recombination graphs under nonrandom ascertainment, with applications to gene mapping. Statistical Applications of Genetics and Molecular Biology 8(1), Article 35.

Björkwall, S., Hössjer, O. and Ohlsson, E. (2009). Nonparametric and parametric bootstrap techniques for arbitrary age-to-age development factor methods in stochastic claims reserving. Scandinavian Actuarial Journal 2009(4), 306-331.

Eriksson, B., Hössjer, O., Järnmalm, K. and Ohlsson, E. (2009). Assessing individual unexplained variation in non-life insurance. ASTIN Bulletin 39(1), 249-273.

Hartman, L., Humphreys, K. and Hössjer, O. (2009). Utilizing identity-by-descent probabilities for genetic fine-mapping in population based samples, via spatial smoothing of haplotype effects. Computational Statistics and Data Analysis 53(5), 1802-1817.

Hössjer, O. (2008). On the coefficient of determination for mixed regression models. Journal of Statistical Planning and Inference. 138, 3022-3038.

Kurbasic, A. and Hössjer, O. (2008). A general method of linkage disequilibrium correction for multipoint linkage and association. Genetic Epidemiology 32, 647-657.

Ängquist, L., Hössjer, O. and Groop, L. (2008). Strategies for conditional two-locus nonparametric linkage analysis. Human Heredity 66, 138-156.

Hartman-Werner, L. and Hössjer, O. (2008). Fast kriging of large data sets with Gaussian Markov random field models. Computational Statistics and Data Analysis, 52(5), 2331-2349.

Sjölander, A., Hartman-Werner, L., Hössjer, O. and Humphreys, K. (2007). Fine mapping of disease genes using tagging SNPs. Annals of Human Genetics, 71(6), 815-827.

Hössjer, O. (2006). Modelling the effect of inbreeding among founders in linkage analysis.  Theoretical Population Biology 70, 146-163.

Kurbasic, A. and Hössjer, O. (2006). Relative risks for general phenotypes. Annals of Human Genetics, 70(6), 907-922.

Bengtsson, H. and Hössjer, O. (2006). Methodological study of affine transformations of gene expression data with proposed normalization method. BMC Bioinformatics 2006, 7:100.

Anevski, D. and Hössjer, O. (2006). A general asymptotic scheme for inference under order restrictions. Annals of Statistics, 34(4), 1874-1930.

Frigyesi, A. and Hössjer, O. (2006). Estimating the parameters of the operational model of pharmacological agonism. Statistics in Medicine 25, 2932-2945.

Hössjer, O. (2005). Spectral decomposition of score functions in linkage analysis. Bernoulli, 1(6), 1093-1113.

Hössjer, O. (2005). Combined association and linkage analysis for general pedigrees and genetic models. Statistical Applications of Genetics and Molecular Biology, 4(1), Article 11.

Johanssson, J-O. and Hössjer, O. (2005). A shot-noise model for paper fibres with  non-uniform orientation. Scandinavian Journal of Statistics, 32, 351-363.

Hössjer, O. (2005). Information and effective number of meioses in linkage analysis. Journal of Mathematical Biology, 50(2), 208-232.

Hössjer, O. (2005). Conditional likelihood score functions for mixed models in linkage analysis. Biostatistics, 6, 313-332. Supplementary material at http://biostatistics.oupjournals.org/.

Ängquist, L. and Hössjer, O. (2005). Improving the calculation of statistical significance in genome-wide scans.  Biostatistics, 6(4), 520-538.

Ängquist, L. and Hössjer, O. (2004). Using importance sampling to improve simulation in linkage analysis. Statistical Applications of Genetics and Molecular Biology, 3(1), Article 5.

Kurbasic, A. and Hössjer, O. (2004). On computation of p-values in parametric linkage analysis. Human Heredity, 57, 207-219.

Hössjer, O. (2003). Asymptotic estimation theory of multipoint linkage analysis under perfect marker information. Annals  of Statistics, 31, 1075-1109.

Hössjer, O. (2003). Assessing accuracy in linkage analysis by means of confidence regions. Genetic Epidemiology, 25, 59-72.
 
Hössjer, O. (2003). Determining Inheritance Distributions via Stochastic Penetrances. Journal of the American Statistical Association, 98, 1035-1051.
 
Anevski, D. and Hössjer, O. (2002). Monotone regression and density function estimation at a point of discontinuity.  Journal of Nonparametric Statistics, 14, 279-294.

Frigyesi, A. and Hössjer, O. (2001).  Kernel estimates of dimension spectra for multifractal measures with connections to nonparametric density estimation, Journal of Nonparametric Statistics, 13, 351-395.

Stromberg, A., Hössjer, O. and Hawkins, D. (2000). The least trimmed differences estimator and alternatives. Journal of the American Statistical Association 95, 853-864.

Sköld, M. and Hössjer, O. (1999). On the asymptotic variance of the continuous-time kernel density estimator. Statistics and Probability Letters, 44, 97-106. 

Opsomer, J., Ruppert, D., Wand, M.P., Holst, U. and Hössjer, O. (1999). Kriging with nonparametric variance function estimation. Biometrics,  55, 704-710.

Frigyesi, A. and Hössjer, O. (1998).  A test for singularity, Statistics and Probability Letters, 40, 215-226.

Hössjer, O. (1997). Reqursive U-quantiles. Sequential Analysis, 16, 119-129.

Ruppert, D., Wand, M., Holst, U. and Hössjer O. (1997). Local polynomial variance function estimation. Technometrics, 39, 262-273.

Gustafsson, R., Hössjer, O. and Öberg, T. (1997). Adaptive detection of known signals in additive noise by means of kernel density estimators. IEEE Transactions on Information Theory IT 43, 1192-1204.

Jones, C. and Hössjer, O. (1996). From basic to reduced bias kernel density estimators: links via Taylor series approximations. Journal of Nonparametric Statistics 7, 23-34.
 
Holst, U., Hössjer, O., Björklund, C., Ragnarsson, P. and Edner, H. (1996). Locally weighted least squares kernel regression and statistical evaluation, Environmetrics, 7, 401-416 (with discussion).

Hössjer, O. (1996). Asymptotic bias and variance for a general class of varying bandwidth estimators, Probability Theory and Related Fields, 105, 159-192.

Hössjer, O. (1996). Incomplete generalized L-statistics. Annals of Statistics 24, 2631-2654.

 Hössjer, O., Rousseeuw, P. and Croux, C. (1996). Asymptotic normality of a very robust scale estimate, Statistica Sinica, 6.

Hössjer, O. and Mielniczuk, J. (1995). Delta method for long-range dependent observations. Journal of Nonparametric Statistics, 5, 75-82.

Hössjer, O. and Ruppert, D. (1995). Asymptotics for the transformation kernel density estimator, Annals of Statistics, 23, 1198-1222.

Hössjer, O. (1995). Exact computation of the least trimmed squares estimate in simple linear regression, Computational Statistics and Data Analysis,
19, 265-282.

Rousseuw, P., Croux, C. and Hössjer, O. (1995). Sensitivity functions and numerical analysis of the repeated median slope, Computational Statistics, 10, 71-90.

Hössjer, O., Rousseeuw, P. and Ruts, I. (1995). The repeated median intercept estimator: Influence function and asymptotic normality, Journal of Multivariate Analysis, 52, 45-72.

Hössjer, O. and Holst, U. (1995). On-line density estimators with high efficiency, IEEE Transactions of Information Theory, IT-41, 829-833.

Hössjer, O., Rousseeuw, P. and Croux, C. (1994). Influence function and asymptotic normality of the repeated median slope estimator, Annals of Statistics, 22, 1478-1501.

Hössjer, O., Rousseeuw, P. and Croux, C. (1994). Asymptotics of Generalized S-estimators, Journal of Multivariate Analysis, 51, 148-177.

Hössjer, O. and Croux, C. (1994). Generalizing univariate signed rank statistics for testing and estimating a multivariate location parameter,
Journal of Nonparametric Statistics, 4, 293-308.

Hössjer, O. and Ruppert, D. (1994). Taylor series approximations of transformation kernel density estimators, Journal of Nonparametric Statistics, 4, 165-177.

Hössjer, O. (1994). Rank-based estimates in the linear model with high breakdown point, Journal of the American Statisitcal Association}, 89, 149-158.

Rousseeuw, P., Croux, C.\ and Hössjer, O. (1994). Generalized S-estimators, Journal of the American Statistical Association 89, 1271-1281.

Hössjer, O. and Mettiji, M. (1993). Robust multiple classification of known signals in additive noise - an asymptotic weak signal approach, IEEE Transactions on Information Theory, IT 39, 594-608.

Hössjer, O. (1992). On the optimality of S-estimators,  Statistics and Probability Letters, 14, 413-419.

Hössjer, O. (1991). The change-of-variance function for dependent data, Probability Theory and Related Fields, 90, 447-467.


Some recent research seminars

Asymptotic estimation theory of multipoint linkage analysis under perfect marker information. June 2002: Seminar
Importance sampling for stochastic processes - with applications to linkage analysis, April 2004: Seminar
On multiple testing in bioinformatics and genetics, November 2004: Seminar
Combined association and linkage analysis for general genetic models, March 2005: Seminar
Invariance principles and spectral decomposition in genetics, April 2005: Seminar
Gene mapping using coalescence theory, July 2009: Seminar
Coalescence theory for structured populations with fast migration, April 2011: Seminar

Popular seminars

Matematiska och statistiska metoder för genletning (popular seminar in Swedish for gymnasium students), March 2006: Seminar and assignments
Matematisk statistik och genletning (open lecture in Swedish at the Royal Academy of Sciences), December 2009: Lecture
Coalescence theory and population genetics, (seminar for Dutch mathematics undergraduate students), May 2011: Seminar

Matlab toolbox for linkage analysis

Linkage toolbox is a computer package for simulation and planning in statistical linkage analysis. Based on a number of well commented
m-functions, it allows the user a great flexibility in terms of combining pedigrees, phenotypes, score functions, genetic models and marker data.
Moreover, the user can easliy incorporate new score functions and genetic models by writing new m-functions in the appropriate format.

The basis of the analysis is the conditional inheritance distribution, i.e. the distribution of inheritance vectors given
phenotypes at the disease locus. In this way binary and quantitative phenotypes are treated within a unified framework. Further, a great variety genetic models are allowed for.

The toolbox incorporates exact multipoint linkage analysis as defined by Kruglyak et al (1996), simulation of NPL and LOD scores under both
perfect and imperfect marker data scenaria. NPL analysis is here interpreted in a wide sense: Conditionally on observed phenotypes
(binary and quantitative), the family scores have mean zero and unit variance for perfect marker data under the null hypothesis of no linkage.

A number of other perfomance criteria, such as noncentrality parameter, slope-to-noise ratio, coefficient of relationship,
relative risk, conditional expected lod score and exact expressions for mean value of NPL score function as function of
chromosome location under a dense marker map are included. These are useful e.g. for planning of linkage studies and sampling.

 To use the program, please download the Linux file linkagecopy0602.tar.gz by right clicking at the link with your mouse. When you have copied it into a map of your computer you open it as

tar xzf linkagecopy0602.tar.gz

You will find a number of m-files in the main directory linkagecopy0602. I suggest you start reading readme.txt where the basics of the program
package is explained.

Ola Hössjer
Department of Mathematics
Division of Mathematical Statistics
Stockholm University
S-106 91 Stockholm
Sweden
 olamath.su.se
Updated: March 22, 2006