And how would you even use this model, given that there are no explanations that help you trust where the prediction comes from…
They decompose a time series into trends, seasonality and residuals. That’s what they are actually modelling.
They cannot predict wars in the Middle East influencing inflation unless there is a seasonal pattern(s).
My guess as to how this would work: the machine will first guess from the data alone if this is one of the categories it has already seen/inferred (share prices, google trend cat searches etc.) Then it'll output a plausible completion for the category.
That doesn't seem as if it will work well for any categories outside the training data. I would rather just use either a simple model (ARIMA or whatever) or a theoretically-informed model. But what do I know.
That is, can it use one time series at time X to predict another time series at time X?
Or is this strictly about finding patterns WITHIN a time series.
As they say in appendix 8:
> We create the synthetic data to reflect common time-series patterns using traditional statistical models. We start with four simple times series patterns:
> • Piece-wise linear trends (I), where the number of the piece-wise linear components is randomly chosen between 2 and 8.
> • ARMA(p, q) (II), where 1 ≤ p, q ≤ 8 and the corresponding coefficients are generated from either a multivariate Gaussian or a uniform, then normalized.
> • Seasonal patterns. In particular we create the sine (III) and the cosine (IV) waves of different random periods between 4 and max context length / 2 time-points and time delays.
If there were no such underlying patterns in the class of all time-series data, then even the idea of traditional time-series models would be fundamentally misplaced.
And since this is a transformer model, it also looks for patterns in the problem-specific input data at inference time, just like how the input context to an LLM influences its output's relevance.
Or just search for the James-Stein paradox.
> How can the same model predict egg prices in Italy, and global inflation in a reliable way?
So, predict sign (branch predictors in modern CPUs also use neural networks of sorts), exponent (most probably it changes slowly) and then predict mantissa using Benford's law.
- decomposition: discover a more general form of Fourrier transform to untangle the underlying factors
- memorization: some patterns are recurrent in many domains such as power low
- multitask: exploit cross-domain connections such as weather vs electricity
I always had difficulties with ML and time series, I'll need to try that out.
There is infinitely more entropy in the real world out there than any model can even remotely capture.
The world is not minecraft.
well...
Or other low-dimensional time domain signals?
The problem isn’t domain generalization, it’s that we keep pretending these models have any notion of what the data means.
People ask how one model can understand everything, but that assumes there’s any understanding involved at all.
At some point you have to ask: how much of “forecasting” is actually anything more than curve fitting with better marketing?
How can the same lossy compression algorithm (eg JPG) compress pictures of everything in a reliable way?
The world is chaotic sure, but there are still truths to be found in noisy time series data; saying that the world is too random to be knowable is a bit dismissive, no?
Rigorous understanding of what is over fitting, techniques to avoid it and select the right complexity of the model, etc, are much newer. This is a statistical issue.
My point is that forecasting isn't curve fitting, even thought curve fitting is one element of it.
Text and anything with lots of high frequency components looks terrible
TPUv5e with 16 tensor cores for 2 days for the 200M param model.
Claude reckons this is 60 hours on a 8xA100 rig, so very accessibile compared to LLMs for smaller labs
https://moment-timeseries-foundation-model.github.io/
https://arxiv.org/abs/2403.07815
A friend at work used one to predict when our CEO would post in Slack, which is verry entertaining to see if correct.
So in my opinion it currently falls into a kind of void. If your use case is worth predicting and you put a data scientist on it, you're better off just training cheaper ARIMA models.
it is far more useful for predictions to look for correlations between time series. This is far more complex than looking for correlations in general because most time series trend up or down and therefore correlate.
FWIW— the only sure fire way to win the trade is to buy time and assume both gross incompetence and negligence when it comes action. The only caveat is if the markets tank enough, this administration will signal capitulation before hand, e.g. Trump mildly capitulating on tariffs last April after the markets proceed to relentlessly defecate themselves.
0-DTE options are typically, and for good reason, stupid gambles. But, right now they can’t even be considered gambling, because there’s zero chance of winning. Not just bad odds, but no odds. Again just signaling how truly malicious this admin is and its disdain for anyone and everyone not close to them.
I genuinely want to know. Thank you
However: white noise is where it really struggles. But real pictures of the real world don't look like white noise. Even though in some sense white noise is the most common type of picture a priori.
Similar for real world time series: reality mostly doesn't look like white noise.
Maybe it would be better to train an LLM with various tuning methodologies and make a dedicated ARIMA agent. You throw in data, some metadata and requested window of forecast. Out comes parameters for "optimal" conventional model.
If you're trying to forecast random data, then yes, it's bullshit. Otherwise you have a chance.
If you are talking about granularity of observations, it would depend on what you are trying to predict (the price in an hour or the price in 12 months?) and how quickly you need the prediction (100ms? Tomorrow morning?). If I had infinite data I would use granularity as a hyper parameter and tune that to a level that produced the best test results.
I am for example currently using weekly averages for non-price data forecasting. I could use daily data but weekly is absolutely adequate for this purpose.
New season of scrubs = new war in the middle east.
> Firstly, we need a multilayer perceptron block with residual connections to convert a patch of time-series into a token that can be input to the transformer layers along with positional encodings (PE). > [...] > Secondly, at the other end, an output token from the stacked transformer can be used to predict a longer length of subsequent time-points than the input patch length, i.e., the output patch length can be larger than the input patch length.
Time series in general have none of this kind of structure that's strictly necessary. I'm sure that many real-world sensors typically have some gaussian distribution aspects + noise and/or smoothness and locality types of assumptions that are pretty safe, but presumably that simple stuff is exactly what traditional time-series modelling was exploiting.
Maybe the real question is just what kind of time-series are in the training data, and why do we think whatever implicit structure that is there actually generalizes? I mean, you can see how any training that mixes pictures of dogs and cats with picturing of people could maybe improve drawing hair, detecting hair, or let you draw people AND dogs. It's less clear to me how mixing sensor data / financial data / anything else together could be helpful.
These tools are good at predicting timeseries that are in fact quite predictable. Like insurances will use this to estimate the number of people who will die from cancer in the next year, the year after that, and so on up to 50 years in the future. The model will extrapolate the progresses made in cancer treatment from the current trend, etc. It is a prediction, cause it's still possible that a breakthrough comes in and suddenly people don't die from a certain form of cancer, but generally it should be roughly correct.
Bitcoin prices are a lot more chaotic, influenced by a ton of unrelated events that shape its path a certain way. There is absolutely no certainty that studying the shape of its past evolution will help in any way understand its future evolution.
Of course here I mean by studying its price alone. If you add more information, like who's behind each trend and why, you have a much better sense of what could happen next.
A string of flips is random, but it's very compressible.
In any case, my point was that reality ain't uniformly random. And not only that: pretty much anything you can point your camera at shares enough similarity in their distribution that we pretty much have universal compression algorithms for real world data.
TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.
This open version is not an officially supported Google product.
Latest Model Version: TimesFM 2.5
Archived Model Versions:
v1. You can pip install timesfm==1.3.0 to install an older version of this package to load
them.Added back the covariate support through XReg for TimesFM 2.5.
TimesFM 2.5 is out!
Comparing to TimesFM 2.0, this new 2.5 model:
frequency indicator.Along with the model upgrade we have also upgraded the inference API. This repo will be under construction over the next few weeks to
Clone the repository:
git clone https://github.com/google-research/timesfm.git
cd timesfm
Create a virtual environment and install dependencies using uv:
# Create a virtual environment
uv venv
# Activate the environment
source .venv/bin/activate
# Install the package in editable mode with torch
uv pip install -e .[torch]
# Or with flax
uv pip install -e .[flax]
# Or XReg is needed
uv pip install -e .[xreg]
[Optional] Install your preferred torch / jax backend based on your OS and accelerators
(CPU, GPU, TPU or Apple Silicon).:
import torch
import numpy as np
import timesfm
torch.set_float32_matmul_precision("high")
model = timesfm.TimesFM_2p5_200M_torch.from_pretrained("google/timesfm-2.5-200m-pytorch")
model.compile(
timesfm.ForecastConfig(
max_context=1024,
max_horizon=256,
normalize_inputs=True,
use_continuous_quantile_head=True,
force_flip_invariance=True,
infer_is_positive=True,
fix_quantile_crossing=True,
)
)
point_forecast, quantile_forecast = model.forecast(
horizon=12,
inputs=[
np.linspace(0, 1, 100),
np.sin(np.linspace(0, 20, 67)),
], # Two dummy inputs
)
point_forecast.shape # (2, 12)
quantile_forecast.shape # (2, 12, 10): mean, then 10th to 90th quantiles.
Because many of these have the same underlying causal structures - humans doing things, weather correlations, holidays.
Well studied behavioral stuff like "the stock market takes the stairs up and the elevator down" which is not really captured by "traditional" modelling tools.
I'm sure people will be doing mechanical interpretation on these models to extract what they pattern match for prediction.
Or, you know, maybe they aren't. Thermometers and photon counts are related to weather sometimes, but not holidays. Holidays are related to traffic sensors and to markets, but not Geiger counters.
> Well studied behavioral stuff like "the stock market takes the stairs up and the elevator down" which is not really captured by "traditional" modelling tools.
Prices are the opposite, up like a shot during shocks, falling slowly like a feather. So that particular pattern seems like a great example of over-fitting danger and why you wouldn't expect mixing series of different types to be work very well.
This might be a totall wrong approach, but I think it might make sense to try to model a matched filter based on previous stock selloff/bullrun trigger events, and then see if the it has any predictive ability, likewise the market reaction seems to be usually some sort of delayed impulse-like activity, with the whales reacting quickly, and then a distribution of less savvy investors following up the signal with various delays.
I'm sure other smarter people have explored this approach much more in depth before me.