explains the usage of the and provides background information about the method
Note: 2017-12-11 : Please do not stress this service by uploading and analyzing large data files with sizes in excess of 100 MB.
If you have such a large dataset, please consider restructuring it by including only the variables absolutely necessary for your analysis.
Using the smallest prossible data file will prevent the system from getting short on memory and allows other users to use this ressource as well.
Thank you for your cooperation.
Update: 2015-06-22 : For performance reasons we need to implement a scheduled restart of the MSM program. Please be aware that on 12h00 CE(S)T on Tuesday, Thursday and Saturday , the MSM program will be restarted. Make sure that you have finished your analyses and downloaded any results by then. We apologize for any inconvenience this might cause you and thank you for your understanding.
Dear MSM users. The MSM is still a relatively new method and we are therefore commited to further investigate the strengths and limitations of the MSM. In a recent series of simulation studies aimed at the estimation of population's usual intake distributions we discovered, that the inclusion of covariates into the regression models of the MSM lead under some circumstances ( when residuals of the regression models are non-normally distributed) to results which were not in line with the expected outcome of the simulated distribution. At the current status of the investigation and until this issue can be resolved, we advise you to use caution when planning an MSM analysis which includes covariates. If you plan to do so, please contact our statistics experts Sven Knüppel
or Heiner Boeing
in advance to discuss potential implications on your analysis.
Your MSM Development Team
is characterized by a two-part shrinkage technique applied to residuals of two regression models, one for the positive daily intake data and one for the event of consumption. The method is applicable to nutrient and food intake including episodically consumed foods. Variation in intake that is explained by socio-demographic variables selected in advance is not affected by the method. The additional use of a is possible. The method is suitable to estimate individual usual intake in the case of many repeated measurements and a long time period. The method can handle different distributions of food and nutrient data and constructs proper .
you can calculate the from 24h recall information and supporting data such as food frequency data.