Hi there. I have a contingency table with frequency of food items (rows) measured across 10 years (columns). I wonder what kind of analysis I should use to model the variation in food frequencies with time (preferably in R).
Thanks in advance for your attention.
Well you might try a test of independence to see if there is a relationship between
the rows and columns. This link should help:
https://www.google.com/search?rlz=1C1CHBF_enUS847US847&ei=g2MfXfGmEOmD9PwPtYyRyA4&q=tests+of+independence+for+2+time+series+in+r&oq=tests+of+independence+for+2+time+series+in+r&gs_l=psy-ab.3...107460.118485..122264...0.0..0.113.1470.16j2......0....1..gws-wiz.......0i71j33i10.1sAqIxcE95s
Best, D. Booth
Hello,
You can deal with:
1- A regression models (by checking the relationships between your variables as indicated by Dear David Eugene Booth );
2- A PCA analysis
3- Time series analysis (espicaly a non parametric time series models)
4- ANOVA Analysis ( if you have a factor covriable) ...
...
It depends of your aims of this study.
In this section i examine the two-way contingency tables studied in Independence Testing using the following log-linear regression model:
where all the xij are dummy variables coded to represent categorical variables. In addition, we also consider more complicated models which contain factors consisting of interactions between the same variables, as described in the sections listed below, and the yi are used to express the frequency of outcomes.
What question(s) do you want to study? Ex. (1) Did counts increase over time? (2) If this items counts increase, does another items counts decrease? etc...
Thanks Matthew -> Hypotheses: Do diet change over the years? If so, does it in a predictable pattern? (for instance monotonic increase of one food item in detriment of other items).
I am grateful for the feedback but some answer do not seem to take into account the non independence of the data.
In this section i examine the two-way contingency tables studied in Independence Testing using the following log-linear regression model:
where all the xij are dummy variables coded to represent categorical variables. In addition, we also consider more complicated models which contain factors consisting of interactions between the same variables, as described in the sections listed below, and the yi are used to express the frequency of outcomes.@ Juan Antonio Balbuena
- Badges
- Science topic
- Similar topics
- Analytical Chemistry
- Column