NZX AI forecast: can machine learning predict New Zealand stocks?

I built Trading Agent in Auckland. So the question I get asked most by other Kiwis is the simplest one: does any of this work on the NZX?

It's a fair question, and an underserved one. Search for "NZX AI forecast" or "New Zealand stock prediction" and you'll find almost nothing — because every global AI-stock tool has made the same business decision: New Zealand is too small to bother engineering for. The NZX's entire market cap is a rounding error next to a single mega-cap on the Nasdaq.

We made the opposite decision. Here's the honest picture of what machine learning can and can't do on our home market.

The NZX is a genuinely different animal

Before any numbers, you need to understand why the NZX is harder to model than the S&P 500 — because the reasons are structural, not just "less data."

1. Liquidity

Many NZX names trade thinly. A handful of large caps — the likes of Fisher & Paykel Healthcare, Auckland International Airport, Mainfreight, Spark, Meridian — carry most of the liquidity, and below them the order books get thin fast. Thin liquidity means wider spreads, larger gaps, and price moves driven by single large orders rather than the smooth, continuous flow a technical model assumes. A model trained on the deep, liquid US tape sees a much noisier signal here.

2. A small, correlated index

The NZX 50 is dominated by a few sectors — utilities, healthcare, property, and a heavy weighting toward defensive, dividend-paying names. That makes the index behave differently from a growth-heavy US market, and it means individual-stock moves are often driven by interest-rate expectations and dividend dynamics rather than the momentum and volatility patterns technical features capture best.

3. Offshore-driven flow

A large share of NZX activity is institutional and offshore. When a global fund rebalances its Australasian allocation, an NZX large-cap can move on flow that has nothing to do with the company or its chart. Those moves are, almost by definition, unpredictable from price history alone.

4. Data and corporate actions

Free data feeds handle NZX corporate actions — imputation credits, frequent capital raises, dividend reinvestment — less cleanly than US data. As with Taiwan, some apparent model error is really data-hygiene noise we're still working through.

The honest numbers (with a giant caveat)

Here's where I have to be straight with you in a way most tools never are: our NZX sample is small.

As of writing, our public verified log has only around 27 verified NZX predictions, running at roughly 52% directional accuracy. That's a touch above a coin flip — but on 27 data points, that number is statistically almost meaningless. It could be 40% or 65% once we have a few hundred verified calls. Anyone who quotes you a confident accuracy figure on a sample that small is fooling you or themselves.

So the honest answer to "does ML work on the NZX?" is: we don't have enough verified data to claim it does, and we won't pretend otherwise. What we have is an early, growing, fully public track record that you can watch accumulate in real time.

Compare that to our US large-cap number — about 54% on ~830 verified predictions — where the sample is large enough to mean something. The contrast is the point: a number is only as trustworthy as the sample behind it, and we show you both the number and the sample on every market.

Why bother with the NZX at all?

If the sample's too small to make claims, why cover it?

1. Somebody should. Kiwi retail investors — and there are a lot of them now, post-Sharesies — deserve a research tool that at least tries their market honestly, instead of pretending it doesn't exist.
2. The track record only grows by running it. The only way to ever have a meaningful NZX number is to start making timestamped, verifiable predictions now and let them accumulate honestly. That's what we're doing.
3. It's our home market. A tool built in New Zealand that ignored the NZX would be a bit embarrassing.

What I'd actually tell a Kiwi investor

Use the NZX output as context, not gospel. A model read on Fisher & Paykel Healthcare with an honest "small sample, ~52%, low confidence" label tells you something useful: this is a thin, offshore-influenced market where our patterns are weak, so weight your own research more heavily than our number. That's a legitimate, honest use of a model that isn't pretending to be an oracle.

For the markets where our model currently has a real, sample-backed edge, that's US large-caps — see the methodology and the full per-market breakdown at /predictions. And if our NZX number firms up as the sample grows, you'll watch it happen live, with every win and loss on the record.

That's the deal: no hype, no hidden losses, and an honest "we don't know yet" when that's the truth.

---

This article is educational content about machine learning and market structure. It is not financial advice, not a recommendation to buy or sell any NZX-listed or other security, and not directed at any individual's circumstances. Trading Agent is a quantitative research tool operated by WU Capital Limited (New Zealand).