National Repository of Grey Literature 9 records found  Search took 0.00 seconds. 
Analysis of brand internal cannibalization due to promo mechanics
Böhm, David ; Reisnerová, Soňa (advisor) ; Dvořák, Marek (referee)
The aim of this thesis is to compare various promo mechanics taking into consideration their behaviour to the same non-promo product or a group of products (cannibalization). The sales volume increase in promo weeks compared to non-promo weeks is the second studied quantity (uplift). The analysis is calculated on the data acquired from the sales of biscuits in 2007 and 2008. The data show sales volumes in separate weeks and markets divided by applied promo mechanics (Bonus Pack, Limited edition, Multipack, Upsize, Temporary price reduction or without promo mechanics) and estimated sale volume that would be sold without promo mechanics (baseline). Products are sorted into categories according to the product properties, price and sales volume in non-promo weeks. The promo mechanics' behaviour is studied on these categories. The results of the analysis will be used as a base for future promotion strategy. The end of this thesis deals with approximation of the distribution function of cannibalization and uplift and reciprocal relation of these quantities.
Statistical methods of the analysis of compound point processes
Zachařová, Tereza ; Reisnerová, Soňa (referee) ; Volf, Petr (advisor)
The risk theory studies mainly the behaviour of compound point processes and processes derived, where in random times increments of random size occur. The main objective of the present thesis is to collect in a systematic way the results on compound point processes and verify them by simulations. The essential parts of this work deal with risk processes and so called ruin event. We concentrate mostly to the case of compound Poisson process, with independent and identically distributed increments. The results concerning both light-tailed and heavy-tailed distributions are presented. To this end, the classification of probability distributions along their tails is recalled, too.
Bayesovská detekce změny v míře nezaměstnanosti s použitím MCMC metod
Reisnerová, Soňa
The time series of numbers of new unemployed is analyzed and modeled as a series of Poisson counts. Its intensity parameter is assumed to have a linear shape with a possible jump and trend change. The change is detected with the Markov chain Monte Carlo procedure. An application to Czech Republic data 1998-2003 is presented.
Analýza vývoje nezaměstnanosti pomocí diskrétního časo-prostorového procesu
Reisnerová, Soňa ; Volf, Petr
The article deals with the discretized version of spatial-temporal random point process model and uses it for the analysis of number of unemployed people. The real components are given by the districts of the Czech Republic and are tied together by a simple model of M.R.F. The temporal components develop as s time series, time runs in months through years 2000-2005. We distingvish between the effects of years and the seasonal (months) effects. The solution uses Bayes approach with MCMC computation.

Interested in being notified about new results for this query?
Subscribe to the RSS feed.