Walk-forward validation let's us test a strategy on a much longer timeframe. In the example below we were able to test a strategy over a very long period of 20+ years. If you want to go through the full post with the code used, go to my substack

he example above represents an unanchored approach to walk-forward validation. There is also the anchored approach whereby the dataset of each training period gets longer as we progress through the iterations. Anchored means that the starting point of all periods is the same as

What is Walk-Forward Validation? It utilizes a rolling window approach where the model is trained and tested on consecutive periods. It creates a situation where the model is updated continuously, mimicking a process we would realistically take when trading this strategy. The

You can check out my Substack for the full description of this trade, as well as some additional ideas on how to make it more profitable.

Finally, the table below shows the summary of the investment performance. To purchase the bond we convert 130,631 USD to EUR at a rate of 1.0575 and sell a single December 2024 futures contract at 1.07965. After one year we sell the bond and convert the entire EUR balance

We will choose a Supranational bond of European Union EU 2 ⅞ 10/20/25 Corp (ISIN: EU000A1A2KZ4) as our investment. The credit risk of the bond is minimal and it has the highest AAA rating from Moody’s and S&P rating agencies.

Now let’s take a look at a real-world example of this trade. Let’s say that we decided to enter this trade on 31 October 2023 and held it for exactly one year, exiting the trade on 31 October 2024. We will use the Euro futures contract that expires in December 2024. On the chart

So why are 1-year futures contracts trading at higher levels than the spot market? The answer comes down to the interest rate differentials between the two currencies involved and the time remaining until the maturity of the futures contract. Forward rates incorporate a premium.

Is it possible to get over 10% return from a bond without too much risk? The investment idea in question provides an investor whose base currency is USD a way to achieve some additional yield on their cash without a significant increase in credit risk. Here is the summary of the

Bank of Japan's Uchida said today that they won't raise rates when the market is unstable. I am very impressed by BOJ's speedrun of their hiking cycle, this must be a new world record.

The real value of these strategies comes when looking at the drawdowns of both strategies: both are musch lower than our Buy-and-Hold benchmark. The Substack post has all the code and commentary, check it out!

The backtest below is also promising: both the Clustering and Random Forest strategies outperform the Buy-and-Hold approach, although both lag it over some periods.

First of all, the graph below shows when our Clustering strategy tells us to stay long in the stock (pink dots) or stay out of it (blue dots). It looks pretty good, avoiding some of the corrections and participating in uptrends.

In this post I explore the use of a Random Forest model to beat a Buy-and-hold benchmark for Microsoft stock $MSFT. Check out my free substack for full code and a detailed overview!

So what is the takeaway here? This doesn't change much, in my view. It only reinforces the idea that rate cuts won't come before June at the earliest. I don't think this print changes much for Fed either. Here is how I imagine Powell looks at today's data

MoM Core CPI came in at 0.44%, also slightly higher than expected. Just like in January, the main culprit is Shelter: this month's reading isn't quite as high as January's, but still higher than expected #CPI #investing #StockMarket

Happy CPI day! YoY CPi comes in at 3.15%, slightly higher than expected 3.1% #inflation #CPI

Rates are down today due to weaker ADP Employment report, continued $NYCB troubles and Powell reiterating for the thousandth time that the economy remains strong with a rate cut coming later in the year #stocks

Today's ADP Employment report came under expectations with only 140k jobs added vs. expectations of 150k. As usual, most jobs were added in services sector, no surprises there #investing #stockmarkets

You can expect a huge pop after trading resumes, since it got the capital injection it needed. I doubt this is the last time it causes jitters in the market, though #investing.