Doing math as an analyst; work before school

  1. I pick up my pen and write down what I see on the screen: 600 out of 14000 rows are selected based on the criteria A = 2 (that’s 4%); if I switch over to A = 5, 135 out of 14000 are selected (that’s only 1%) — I now know the probability of several outcomes based on two possible inputs. I do this a few more times and determine the probabilities of several more outcomes based on more inputs, and in the process understand the data much better. I document this so my future self, and others who may use the data, are aware.
  2. Salesperson A should have a overall sales target 10% less than Salesperson B. Given sales targets for multiple products, I have to determine how much to allocate to Salesperson A and how much to allocate to Salesperson B. I make a few notes, writing  down in algebraic terms the relationship between the total sales target, the sales target for person A, and the sales target for person B. I perform rudimentary simultaneous equations and input the formula into Excel, quickly calculating the breakdown for each of the products.
  3. The variable component of a salesperson’s salary is 30%, and  30% of that is determined by subjective factors, input by the line manager; the rest is determined by performance against sales targets. I create a spreadsheet formula that allows the line manager to rate the salesperson from 0-100%, automatically adjusted to fit the weights of the variable component and subjective factors.

I had such a hard time trying to figure out why I needed learn math. That’s why sometimes work before school makes sense.

On theory, practice, and Snowflake Schemas

Just the other day I learnt that the data warehouse I was working on was designed using a Star and Snowflake schema. I’d known enough about them to know that this meant the data was set up on fact and “dimensional” tables, but not much other than that.

So the moment I had some time I went online and looked up definitions, and realised that they were pretty much the way many of the bigger databases I’d done up looked like. I’d been using this schema for the past four years (at least) without my ever realising it. It was like in Le Bourgeois gentilhomme when Jourdain remarks

Good heavens! For more than forty years I have been speaking prose without knowing it.

Which reminded me of something else I read about in Taleb‘s Antifragile: that academia often comes after practice. You can do something your whole life (practice), have it labelled and described as something in theory (academia), and after a while forget that it’d started not as a theory but as an unlabelled, undescribed bit of practice and not as a theory.

Personally, Taleb’s spelling this out gives me much reassurance that we don’t always have to understand something theoretically in order to do something (an activity) well. If I want to run well, or swim well, or program well, it doesn’t necessarily have to follow on from learning the theories of aerodynamics, water viscosity, or binary.

Student loans and how the deed is infinitely stronger than the word

Interesting article giving the perspectives of three people with outstanding student loans and how they’re paying it off.

I’d never been that heavily in debt and I do sometimes wonder what I’d do. Though I cannot say for sure, I do not see myself holding off the payment of loans if I could I help it. But I probably won’t need to theorise much more as in a few years time my mortgage will kick in and it’d be interesting to see if I’d practice what I preach about paying loans off as soon as I can. I cannot say for sure if that’s what I’d do.

That’s the thing about people: they may foresee themselves doing one thing, and saying they’d do it, only to do something else altogether when they really do it. They’d say I’d buy Brand X, definitely and then just as quick go on to buy Brand Y for no other reason than that because they felt like it. That’s definitely something to think about when collecting user responses in surveys and the like when major decisions are going to be made based off of it.

Coursera

I just found out about Coursera last week. Yes, I know, I’m late to the party! If you’re late like me, here’s what’s Coursera about (taken from their About page):

We are a social entrepreneurship company that partners with the top universities in the world to offer courses online for anyone to take, for free.

Yes, I know. I drooled like you. The possibilities are endless!

I signed up for a never of courses myself, mostly on business and analytics, though I’m thinking of dabbling in some “outside the box” stuff. These are exciting times.

On Ignorance and Information Search

I’m not a person who takes not knowing lightly.

If someone asks me a question and I’m unable to find the answer off the top of my head, chances are good that within the next few minutes, armed with a computer and a good internet connection, I’ll find the answer to that question. Of course this is assuming it’s a question that intrigues me enough for me to do so (but then again simply not knowing something often intrigues me enough to push me to find its answer).

I find ignorance a chosen state; in general, people do not not know something not because they’re stupid, but because they’ve never had a need or want to find out what that something is. Motivation’s seriously understated in education. Teach a man to fish, and if he’s not hungry he may not learn. Teach a man hungry to learn to fish, well, that’s another story.

For me it’s not so important to remember any of the actual facts that I’ve looked up as much as it is knowing how to look up that fact in the first place. For example, I have no idea what’s pi to its 8th decimal, but I do know that if I searched for pi in Google or Bing I’d be able to find out (it’s 3.14159265).

I can think of at least two reasons for placing teaching how to learn and search for information before teaching facts (something most schools are only too guilty of).

Firstly, information search is so much faster and vastly improved now with the advent of the internet and search engines like Google, making the skill of remembering lots of facts redundant — you don’t need to remember a fact you only need to use once or twice as searching it up may well be faster than the time it takes to burn it into memory; and secondly, because many facts in life are dynamic and may have changed since you first learned it (e.g. when I was in school Singapore’s population was at about 3.3 million; it’s over 4 million now, and may have changed by the time you read this).

I’ve always harboured a slight distrust of people who utter the words “I don’t know”, and will always be wondering at the back of my mind whether or not that person had attempted to find an answer. I urge you to never use those words unless absolutely necessary (e.g. if your child asks you what your neighbours were doing in the back of that shaking car). If possible, always answer a question to which you don’t know the answer to with a “I’ll find out” or “I’ll get back to you on that”. And do.

As for what WERE the neighbours doing… I’ll get back to you on that.