Financial Statistics

News

• The doodle for the seminars is here


• The exercise/question time on Thursday is moved to Monday 15-17 in MH:330
• Dr. Jonas Persson from Sungard is giving a guest lecture today.
• The projects have been handed out. All seminar will be in MH:330.
• Feel free to sign up for computer exercise 4
• Feel free to sign up for computer exercise 3
• I have moved/rebooked all lectures in M:D to MH:309a
• Feel free to sign up for computer exercise 2
• 6/11 The lecture on Thursday 8/11 is moved to M:D.
• 5/11 The course program was updated (computer exercise 2 on Wednesday 14/11 was moved to Tuesday 13/11)
• Feel free to sign up for computer exercise 1

Lectures

• Summary lecture here.
• Lecture 12 (including applications) here
• Paper on: Particle filters and Bayesian inference in financial econometrics
• More on particle filters IntroductionToSMCmethods.ps
• Lecture 11 here
• Paper on EKFs and calibration of options prices
• Papers on parameter estimation Fokker-Planck method Monte Carlo method for diffusions Monte Carlo method for Jump-diffusions and Levy processes Unbiased Monte Carlo Series expansion
• Lecture 7 here
• On non-parametric GARCH models here
• Lecture 6 here
• On portfolio selection with many assets 50% short paper
• Lecture 5 here
• More on GARCH models: Different type of ARCH models Multivariate GARCH models
• Papers on HMMs and spot-forward spread. HMMs and dependence between energy markets .
• Lecture 2 here
• Lecture 1 here
• Introduction lecture: Paper on stylized facts.

Exercises

• Kalman exercise can be found here ExerciseKalman.pdf.
• GMM/EF exercise can be found here ExerciseGMM_EF.pdf.
• SDE exercise can be found here ExerciseSDE.pdf.
• ARCH GARCH exercise can be found here Exercise_Variance_models.pdf.
• Linear time series exercise can be found here ExerciseLinearTimeSeries.pdf.

Computer exercises

• Computer Exercise 4 can be found here: FSlab 4 Data files needed are CIRstruct2.mat, TermStruct.mat and StochVol.mat.
• Computer Exercise 3 can be found here: FSlab 3. Data files needed are regdataN.mat, cirdata.mat and cklsdata.mat.
• Computer Exercise 2 can be found here: FSlab 2 is uploaded. Data files needed are garchdata.mat, egarchdata.mat and SwedishStockData.mat and Matlab routines ccc_mvgarch.p and ccc_mvgarch.m
• Computer exercise 1 can be found here: FSlab1.pdf You need some additional files: regdataN.mat, regdataT.mat, prisdata.mat, sacfplot.m and MLmax.p

Courseprogramme

The preliminary course programme can soon be found here: Courseprogramme.

Literature

Financial Statistics

Course Coordinator/Lecturer

Erik Lindström, tel 046-222 45 78, MH:129.

Teaching assistant

Stefan Adalbjörnsson, tel 046-222 79 76, MH224.

Johan Svärd,

General information

University credits: 7.5 Grading scale: TH Level: A
Language of Instruction: The Course will be given in English

Lectures: Monday 8.15-10.00, M:E, Tuesday 15.15-17.00 E:1406
Exercise: Thursday 10.15-12.00 MH:362C
Computer exercises: MH:230.

The course starts October 24.

Computer exercises

Aim

The course should be regarded as the statistical part of a course package also including TEK180 Financial Valuation and Risk Management and FMS170 Valuation of Derivative Assets. Its purpose is to give the student tools for constructing models for risk valuation and pricing, based on data.

Knowledge and Understanding

For a passing grade the student must:

• handle variance models such as the GARCH family, stochastic volatility, and models use for high-frequency data,
• use basic tool from stochastic calculus: Ito's formula, Girsanov transformation, martingales, Markov processes, filtering,
• use tools for filtering of latent processes, such as Kalman filters and particle filters,
• statistically validate models from some of the above model families.

Skills and Abilities

For a passing grade the student must:

• be able to find suitable stochastic models for financial data,
• work with stochastic calculus for pricing of financial contracts and for transforming models so that data becomes suitable for stochastic modelling,
• understand when and how filtering methods should be applied,
• validate a chosen model in relative and absolute terms,
• solve all parts of a modelling problem using economic and statistical theory (from this course and from other courses) where the solution includes model specification, inference, and model choice,
• present the solution in a written technical report, as well as orally,
• utilise scientific articles within the field and related fields.

Contents

The course deals with model building and estimation in non-linear dynamic stochastic models for financial systems. The models can have continuous or discrete time and the model building concerns determining the model structure as well as estimating possible parameters. Common model classes are, e.g., GARCH models with discrete time or models based on stochastic differential equations in continuous time. The course participants will also meet statistical methods, such as Maximum-likelihood and (generalised) moment methods for parameter estimation, kernel estimation techniques, non-linear filters for filtering and prediction, and particle filter methods. The course also discusses prediction, optimisation, and risk evaluation for systems based on such descriptions.

Required Qualifications

MIO140 Financial Management, FMSF10/FMS045 Stationary stochastic processes or corresponding courses, and preferably also one or several of FMS051 Time series analysis, TEK180 Financial Valuation and Risk Management, and FMS170 Valuation of Derivative Assets.

Literature

Madsen, H, Nielsen, J N, Lindström, E, Baadsgaard, M & Holst, J: Statistics in Finance. IMM, DTU, Lyngby and KFSigma, Lund 2006.
Additional handouts

Assessment

Written report and oral presentation of a larger project and compulsory computer exercises. The course grade is based on the project grade.