Not at all. Actually, you’re early!

Less than two percent of the world own cryptocurrencies. This nascency in adoption is a main reason individuals can make so much money in this space right now. Call it first mover’s advantage.

Early crypto adopters have faced a high barrier of entry navigating the space. It can take a certain degree of finance and tech savvy — with a healthy dose of suspicion for government tomfoolery — to get involved.

As a technology, blockchain — and associated cryptocurrencies that lubricate their networks — form part of the next generation of disruptive technologies terraforming…

Ten years ago, if you put aside $4 for that cup of Starbucks tall coffee into Bitcoin, you’d be sitting on a whopping $688,050 today. Insane right?

That’s an annualised return of 234%. Compare that with:

• stocks (10.5% for the S&P 500 index),
• real estate (3.8% for US median house price) and
• bonds (0.7% for the Vanguard BND Total Bond Market ETF).

Albert Einstein is reputed to have said:

“Compound interest is the eighth wonder of the world. He who understands it, earns it — he who doesn’t, pays it.”

Appreciating the life-changing power of compound interest can transform your life, whether you’re scaling your startup or investing for financial independence.

2021 Jun update: I have published sequel articles for the crypto space here and here, and began discussing high-compounding investments on YouTube.

1. Compound Interest = Exponential Growth

Many people mistake compound interest for simple interest, which grows your assets linearly. Compounding assets grow exponentially*, which mathematically skyrockets their value. This effect is dramatically enhanced by the…

Let’s compare and contrast differential equations (DE) to data-driven approaches like machine learning (ML).

In a nutshell — albeit with caveats — they can be thought of different approaches to modelling various phenomena. Do you make the rules? Or should you let your juicy data learn the rules for you?

Both types of models absolutely drive the world around us. Let’s dig in.

Edit: I have written a sequel article here.

Example DE models

Navier-Stokes (meteorology)

The model behind weather predictions. It is a chaotic model — meaning predictions can be wildly off when using just slightly incorrect inputs. That’s why weather predictions are often…

A hallmark of cryptocurrency bull runs is retail FOMO. New investors enter the space en-masse and speculate into various tokens toward peak prices. To complicate matters, the 2021 bull run saw the explosion of leveraged trading.

What’s the result? Loss and disaster.

Pumps always come with dumps. And mass liquidations on leveraged traders cause dumps to be dumpier than they need to be.

Smart money — hedge funds, institutions and whales — then swoop up the bargain bitcoin at the expense of the little guys.

Lessons. Don’t FOMO. Don’t buy into resistance. Check your emotions at the door. …

In 2021, the Binance Smart Chain (BSC) established itself as the premier blockchain for high-rewards staking and low gas fees. This double-whammy has generated great excitement among passive income crypto investors.

In this article, we’ll do a mathematical deep dive of some top passive income protocols on the BSC. These shall include single-token staking stables like CAKE (110% APY) and BUNNY (225%), along with some even-higher returning dual-token staking LPs like HOTCROSS-BNB (849%) and the ridiculous WINGS-BNB (12,000%+). For reference, the staple passive income vehicle of the stock market are dividend shares, which return 1–10% per year.

This is Part…

Hi,

In the spirit of responsible data science, I think this article should be called something along the lines of "Attempting to predict bitcoin prices with ML -- can it be done?", with a section at the end dedicated to why the answer is a big "no". Similar story for predicting Dogecoin or stock prices.

Equity pricing is one of those use cases that simply doesn't suit ML. You're looking at prices that essentially operate as random walks. Thus you can't predict these prices based on its own values with any meaningful accuracy. This has been shown time and time…

This article is Part I of an series on deep-diving how machine learning algorithms are evaluated.

Here, we’ll visually review the most popular supervised learning metrics for

• Classification — Accuracy, Precision, Recall, Fᵦ & AUC; and
• Regression — MAE, MSE and R².

In short, the more advanced classification metrics allow you to calibrate the importance of Type I and II errors for your use case, while dealing with imbalanced datasets. We’ll also visually explore some connections between classification metrics and probability.

In this and this article, I apply these classification and regression metrics to real-world ML problems in Jupyter using…