relative time windows
interesting idea - not relevant to our study
relative time windows
interesting idea - not relevant to our study
beyond prediction and into the realm of causality can mislead
what does this mean in practice?
Overall, the available evidence supports that species-specific C–N–P stoichiometry is determined by multiple traits and environmental factors
this is the overall idea I want to get from the paper I think
ratios
wow
Parameterizing
estimating parameters
LANDIS-II
do you use LANDIS?
reconstruct disturbances over a longer period of time, using a variety of data sources such as archived satellite imagery, air photos, dendrochronology, and disturbance records
Bo's?
Strengths and weaknesses of alternative empirical approaches for investigating disturbance interactions (DI).
ours serendipitous (SBW) and experimental (moose exclosure)
Filename
no data or code included
that weather effects (either post-fire or recent antecedent) were more important for predicting post-fire perennial and annual grass cover at a new site than they were for explaining variability at a single site
interesting example of model for inference may be very different than model for prediciton
nor did it reveal that weather during specific post-fire recruitment periods
appreciate that their study had parts where tehre was no lasting effect and it was tsill published
pre-selected weather variables
why would you put this here?
μ
these dont show up great in this format
a modified forward stepwise model selection procedure
is this just a bottom up approach? maybe not bottom up but starting low and going upwards in complexity
a priori hypotheses of the factors influencing lichen presence
would have liked if they reiterated what those were
varIdent
?
a generalized linear model, identified stand age, stand type, and ecoprovince as significant predictors of lichen presence. With a logistic growth model, a measure of lichen recovery (time to 50% asymptotic value) varied from 28 to 73 yr
is this doing exploratory and inference at the same time?
simultaneously
not simultaneously but needs to be sequential. if wanting to to predictions you would need models established through inference but calibrated/validated on part of your data set
cross-validate
always works
one
I think gone beyond exploratory for this project - previous work has established relationships
Changing the data collection strategy in response to forecasts is a unique opportunity that forecasting offers in comparison with retrospective analysis, and is transformative as a means of distinguishing between competing hypotheses
very platt's method