Course Map: From Reading Data to Modeling It

Welcome to the second course of the institute.

In SVI 101 you learned to read the market honestly — to look at returns, distributions, and the numbers Saqrid puts on screen, and to ask whether any of it is genuine evidence. That course was about not being fooled by data. This one is the other half: the probability that produces the data in the first place. If 101 taught you to be a skeptical reader, 102 teaches you to be a modeler — to build the machinery the platform's outputs are computed from, so that when you read a probability tunnel or a volatility band, you know exactly what assumptions are humming underneath it.

A fair warning, and an invitation: 102 is more formal than 101. You will see real equations here — densities, expectations, a covariance matrix, a stochastic differential equation or two. Do not let the symbols intimidate you. Every one of them is introduced from intuition first, then a picture, then a worked number, and BaPig is in every chapter when a line does not land. The math is not decoration; it is the difference between using a tool and understanding it.

Here is the road through the six modules, and the Saqrid output each one is built to explain:

1. Random Variables & Probability Foundations — the formal language of an uncertain quantity: discrete versus continuous, the density and the CDF, expectation and variance as operators, and the distribution catalogue. This is the grammar everything else is written in.
2. The Normal & Lognormal Return Model — the model that SVQ's probability tunnel and SV's Normal Cone actually draw: normal log-returns, lognormal prices, geometric Brownian motion, and the volatility drag.
3. Conditional Probability, Bayes & Expectation — how a belief updates when you learn something, and the conditional expectation that the SVQ cone is a stack of. This is also where the seed of regression is planted.
4. Measuring Historical Volatility — the three estimators behind SV's HV(P&R): close-to-close, Parkinson, and Garman-Klass, and why using more of each price bar gives a sharper number.
5. Smoothing, Filtering & Trend Extraction — smoothing treated as statistical signal extraction, not retail moving-average tricks: the filter view, the lag-versus-noise tradeoff, and the EWMA variance estimation that real volatility tools rely on.
6. Portfolio Mathematics — how risk lives not in any single asset but in the covariance matrix of how they all move together: diversification, the efficient frontier, and what powers SVQ's mean-variance optimizer.

The course ends with a graded assessment that runs the whole toolkit at once. Pass it and 102 counts toward your Tier 100 credential; earn honors and you are flagged for the honors track.

When you finish here, you will have the complete probabilistic foundation — and Tier 200 takes it live: the inference, regression, and GARCH volatility models that turn these clean assumptions into things you can actually estimate, test, and trade.

Let's begin where all of probability begins — with a single number you do not yet know.