People's behavior is at least partially determined by the information that they have/are receiving. [Behavior as a function of information + other variables]
That a researcher is also looking for PDF-to-text tools for text analysis: https://askubuntu.com/a/344080
Calibre might offer the best, general method of extracting text from PDFs.
Relative environmental costs/benefits (social? ScopeOfApplication between Social/Private in Economics, EnvironmentalEcon) in terms of freezing veggies to lock in nutrients versus canning [it was suggested to me by a friend that canning removes nutrients from vegetables whereas freezing them does not]. However, is the benefit conferred to an individual that consumes frozen veggies, from the added nutrients, worth the cost of maintaining the frozen state of veggies? [Could change according to changes in the efficiency of freezing and temperature control technology, such as passive heaters/coolers that are well insulated via vacuum technology; the initial canning process may be more energy-intensive than freezing and maintenance of a frozen state]
Discussion(s) above entails the consideration of functions and variables involved with them. It would be nice to have some manner of perceiving and updating connections between these. [ProjectIdeas, Ideas, FunctionalNotation]
Looking into Logit and Probit statistical methods! ...And how they differ from linear regressions! https://stats.stackexchange.com/a/30909 talks about the meat of linear models. + there's other insightful responses to the question that prompted it ("Difference between logit and probit models").
Did you know that "Probit" means probability unit? Does "Logit" then mean logical unit? Apparently not; rather, it deals with logarithms: "the logit function gives the log-odds, or the logarithm of the odds" (Wiki article).
I'm looking into this for a couple of reasons. Firstly because Ryan and I were recently talking about it. Secondly, I haven't been exposed to much non-linear regression thinking/methodology. I'm aware of the case in which one determines the probability that a dependent variable is
True given various independent variables, which I think is Probit, but I'm uncertain. And I'm quite certain that this would be a good/useful thing to know. Thirdly, I saw the post "Linear Regression in Python" on Planet Python, which reminded me that I should delve into learning something that doesn't involve linear regressions.
Oh and on the DataFramed podcast, whose archives I've been listening to, the speakers have mentioned a couple of times that using "logistic regression" ("Logit," right?) is often either sufficient enough for many applications of statistics/DataScience within business versus MachineLearning or it's more efficient/marginally insightful.
On the mention of "marginally" above, I think that I'll now make a page for MarginalAnalysis that I can link to when I mean to use the word "marginal" in the way that the term is used in economics. Ryan and I were discussing that it can be very confusing to people that are more familiar with the conventional usage of the term "marginal", which suggests that a given thing — the marginal thing — is insignificant or inconsequential. In economics, *marginal* talks about the addition of one unit of something to a whole. For example, the addition of the integer one (
1) to a different integer one (
1) leaves you with a group that includes two ones. [Funny, I had attempted to use this to discuss rates of change via adding one, but in attempting to structure this statement I'm wondering again about how numbers "work". I've here "formed" a set that has a "length" of two, as it includes two "objects" (can you tell that my time spent with Python has changed me?). But when I'm attempting to add the integers within the set (
1) together, I appear to be doing something different than simply getting the length of the set. (Perhaps I'm mostly being confused and amused by that in this case the length of the set happens to equal what would also be the sum of the integers contained within the set.) But the addition seems to involve some additional process being done over the set, or maybe conducting some "operation" or behavior that involves all of the objects/elements/items contained within the set in question.]
Candidacy for refactoring within a Wiki system, or any system that involves a notion of canonical content. How to determine this?
Data as asset. Data as liability.