This is a mobile optimized page that loads fast, if you want to load the real page, click this text.

Dump it Here

Jump to new

The article by Chris Conway of Livewire Markets, which inspired this exercise, highlights the evolving capabilities of (AI) in stock picking. As (AI) technology continues to improve, it will be interesting to see how it compares to humans in the field of finance.

Explanation - Fundie vs AI: The best ASX "Income" stocks for the next 12 months
This is a theoretical investment exercise pitting Google's (AI) Gemini against Dr Don Hamson from Plato Investment Management in selecting five "Income" stocks for the next 12 months.

Fundie vs AI: Who is the better stock picker? (and 10 ASX stock ideas for your trouble)

AI is getting better, faster, & stronger. Will it ever be able to pick stocks like a human? We take a step towards answering that question.
www.livewiremarkets.com

Results after 32 weeks
We recognise that this is nothing more than a fun, but nonetheless important, thought experiment. Ultimately, (AI) will likely become a tool that all use to make more informed decisions, across a broad range of topics.





Skate.
 

"Fundie vs AI: Who is the better stock picker?"
Yesterday's report focused on "income" stocks. Today's post examines "growth" stocks.

The article by Chris Conway of Livewire Markets
The article inspired this exercise, which highlights the evolving capabilities of (AI) in stock picking. As (AI) technology continues to improve, it will be interesting to see how it compares to humans in finance.

Explanation - Fundie vs AI: The best ASX "Growth" stocks for the next 12 months
This is a theoretical investment exercise pitting Google's (AI) Gemini against Tobias Yao from Wilson Asset Management in selecting five "Growth" stocks for the next 12 months.

Fundie vs AI: The best ASX growth stocks for the next 12 months

We've seen the income stocks, now we're looking at growth stocks in this edition of "Fundie vs AI: Who is the better stock picker?"
www.livewiremarkets.com

Results after 32 weeks
We recognise this is nothing more than a fun, but important, "thought experiment" as ultimately, (AI) will likely become a tool that all use to make more informed decisions, across a broad range of topics.



Skate.
 

360 was a good pick. Did the expert use AI to pick that?
 

Fundie vs AI: Who is the Better Stock Picker?
Some use #Google and others use #Microsoft. Adding Microsoft (AI) into the mix adds another layer of comparison. Today’s second post delves into the world of “growth” stocks by comparing the picks of two (AI) models and an Expert Funds Manager.

Inspired by a recent article
This exercise highlights the evolving capabilities of AI in stock picking. As AI technology continues to advance, it will be fascinating to see how it stacks up against human expertise in finance.

Explanation - Fundie vs AI: The Best ASX Growth Stocks for the Next 12 Months
This theoretical investment exercise pits Google’s (AI), Gemini, and Microsoft’s Copilot (AI) against Tobias Yao from Wilson Asset Management. Each contender has selected five growth stocks from the ASX for the next 12 months.

The Results Demonstrate the Direct Comparison
While this is primarily a fun and insightful “thought experiment,” it underscores an important point - "different AI models" vary in their stock picks based on the data they draw from. Going forward, (AI) is likely to become a valuable tool for making more informed decisions across a broad range of topics, including the world of trading.



Skate.
 
Last edited:
Sean K said:
360 was a good pick. Did the expert use AI to pick that?
Click to expand...

@Sean K, I’m sure Tobias Yao from Wilson Asset Management has access to a wealth of information beyond what we can imagine.

Wilson Asset Management likely selects growth stocks through a detailed analysis of financial metrics, market trends, and industry positions. These money managers focus on companies with strong sales and EPS growth, high ROE, positive free cash flow, and favourable market conditions to identify stocks with significant growth potential.

This analysis underscores the critical role of stock selection and market conditions in determining long-term portfolio performance.

Skate.
 

Just Thinking Out Loud
Using Artificial Intelligence (AI) for stock selection offers significant advantages by leveraging detailed analysis to identify strong buying opportunities and favourable market conditions. (AI) can pinpoint stocks with potential, making both buying and selling decisions based on comprehensive data analysis. This approach removes emotional biases from trading, as (AI) operates on information that might be unavailable or not commonly known to the public.

Effective trading
Hinges on stock selection and market timing to determine portfolio performance. By relying on Artificial Intelligence (AI), traders can benefit from these 4 points:

#1. Unbiased Decision-Making
Artificial Intelligence (AI) makes decisions based on data, not emotions.

#2. Access to Advanced Analysis
Artificial Intelligence (AI) can process vast amounts of data quickly, identifying trends and opportunities that might be missed by human traders.

#3. Cost Efficiency
Using Artificial Intelligence (AI) reduces the need for extensive trading experience and costly data and software.

#4. Future-Proofing
Artificial Intelligence (AI) is likely the future of trading. Major trading firms have already integrated (AI) to sort and analyse the data that @Sean K mentioned.

My Question
Could Artificial Intelligence (AI) be the answer for us as well to enhance our trading?

I would be interested in @Richard Dale input as he is an expert in the field.

Skate.
 
An "AI" in this context, refers to a Large Language Model (LLM) that utilizes a set of sampled data to generate responses to queries. To a layperson, these responses may seem intelligent because they are often coherent and contextually appropriate.

However, when LLMs are applied to investing, they often lead to exaggerated claims that these "AIs" can deliver higher trading or investment returns than humans.

In reality, an LLM, particularly in financial scenarios, functions as a probabilistic simulation engine. In simple terms, it generates the response with the highest likelihood based on its training data. It lacks foresight and does not truly understand the questions it is asked, which is a significant limitation of LLMs.

Asking any AI engine about which stocks to choose is probably no better than a dartboard. If a new set of data is added to the intial data set, the results will change (without notice) and certainly with no reason. This might be a blog posts extolling the virtues of X, Y or Z industries. They will certainly change the tokens used and probabilities therein, but it doesn't mean that anything has actually been improved.

"AI" certainly has its uses in providing reasonable probability to factual questions, given enough high quality input data, but the predictive value is questionable.

Any use of such in a future prediction of returns in an investing portfolio is basically nonsense, and, furthermore, a
sample size of 1 is anecdotal and lacking in any statistical significance. Skate, this is your Gemini model.

Machine Learning (ML) and Large Language Models (LLMs) excel at analysing and offering insights into evolving trends in market commentary. However, this sophisticated application of ML/LLMs surpasses the knowledge of those who simply input queries into an "AI" engine, hoping it will act as a guru to predict future returns. The AI portion of this thread is mostly that - you are just substituting one dart board for another with little to no understanding of your dart's trajectory or even the dart board "sample" available to you.

Put another way, let's say you are a Search Engine company going into "AI" and the majority of the announcements/commentary that you sample is on Biotech companies. What do you think this engine is going to produce? What happens when the model expands to incorporate a big stack of Rare Earth Mineral companies. Both types of companies would have all sorts of emotive future-looking language. Contrast this with a Utility or Infrastructure company that has a (mostly) steady model without much variation. Do you think your AI/LLM is ever going to suggest that a (boring) Utility company as a high probability match for anything? Nope!

But maybe boring is fine:

 

@Richard Dale, thank you for your detailed response. You’ve highlighted some important limitations of (AI) when it comes to trading as an (AI) model can’t truly understand or predict market movements with certainty or accuracy.

(AI) Potential
However, I believe the future of (AI) holds immense potential, especially when used appropriately to analyse vast amounts of data quickly, identify trends, and provide insights that we might miss.

Garbage in, garbage out
I share your scepticism regarding how (AI) gathers its information, however, as an (AI) model, it can answer basic trading questions and evaluate training data to provide simple trading insights.

Here is a Sample question I've posed
(AI), can you provide a list of the top 10 companies to buy with a share price between $0.05 and $15.00 that are constituents of the ASX 200 index? The selection of the 10 companies should be based on the following criteria, aiming to identify companies well-positioned for potential growth and stability in the near term.

#1. Share Price Range
Companies with a share price between $0.05 and $15.00.
#2. ASX 200 Constituents
Companies must belong to the ASX 200 Index.
#3. Sector Diversification
Ensure a mix of companies from different sectors to spread risk and capture opportunities across various industries.
#4. Recent Performance and Market Trends
Companies that have shown strong recent performance or have positive market trends.
#5. Fundamental Strength
Companies with solid fundamentals, such as strong financials, market position, and growth potential.
#6. Growth Potential
Companies with potential for growth based on industry trends, market conditions, and company-specific factors.

Summary
(AI) had no issue with the question or addressing the criteria listed above, and responded based on my input.

Skate.
 

The response from (AI) is RAW and unchecked
Based on the criteria provided, here are 10 companies from the ASX 200 index with a share price between $0.05 and $15.00 that are well-positioned for potential growth and stability in the near term:
  1. AMP Limited (AMP)
    • Sector: Financials
    • Reason: Established financial services company with a strong market presence and solid fundamentals.
  2. Beach Energy Limited (BPT)
    • Sector: Energy
    • Reason: Significant player in the oil and gas industry with growth potential and positive market trends.
  3. Adairs Limited (ADH)
    • Sector: Consumer Goods
    • Reason: Leading retailer in home furnishings with consistent performance and strong financials.
  4. Alumina Limited (AWC)
    • Sector: Materials
    • Reason: Key player in the alumina and aluminium industry with strong fundamentals and growth potential.
  5. Arafura Resources Limited (ARU)
    • Sector: Materials
    • Reason: Engaged in rare earths exploration and development, with potential for growth and positive market trends.
  6. Australian Agricultural Company Limited (AAC)
    • Sector: Consumer Staples
    • Reason: Major player in the beef industry with a strong market position and solid fundamentals.
  7. Bank of Queensland Limited (BOQ)
    • Sector: Financials
    • Reason: Well-established regional bank with solid financial performance and growth potential.
  8. G.U.D. Holdings Ltd (GUD)
    • Sector: Consumer Discretionary
    • Reason: Diversified industrial company with a strong market presence and positive market trends.
  9. Austal Limited (ASB)
    • Sector: Industrials
    • Reason: Leading shipbuilder with a strong order book and growth prospects.
  10. Accent Group Limited (AX1)
    • Sector: Consumer Discretionary
    • Reason: Leading retailer and distributor of footwear with a strong market presence and consistent performance.
These companies have been selected based on their share price range, sector diversification, recent performance, fundamental strength, and growth potential. It’s important to conduct your own research and consider your investment goals and risk tolerance before making any investment decisions. If you have any specific criteria or sectors you’re interested in, let me know, and I can provide more tailored recommendations!

Skate.
 
Another Question for (AI)
(AI), can you provide a list of the top 10 companies to buy with a share price between $0.05 and $15.00 that are constituents of the ASX 200 index? The selection of the 10 companies should be based on the following criteria, aiming to identify companies well-positioned for potential growth and stability in the near term:

AFL Code Posted into (AI) after the question
After asking the question above - I then posted a simple “Enhanced RSI Strategy” code, written in Amibroker AFL, for (AI) to follow to generate buy and sell signals.

(AI) Response
Here’s a list of the top 10 companies to buy with a share price between $0.05 and $15.00 that are constituents of the ASX 200 index, based on the criteria you provided:

#1. Life360 (ASX: 360) - Technology sector, strong recent performance
#2. TPG Telecom (ASX: TPG) - Communications sector, solid fundamentals
#3. SiteMinder (ASX: SDR) - Consumer cyclical sector, positive market trends
#4. Endeavour Group (ASX: EDV) - Consumer defensive sector, growth potential
#5. Woodside Energy (ASX: WDS) - Energy sector, strong financials
#6. AUB Group (ASX: AUB) - Financial services sector, market position
#7. Nanosonics (ASX: NAN) - Healthcare sector, industry trends
#8. Aurizon Holdings (ASX: AZJ) - Industrials sector, recent performance
#9. Lendlease Group (ASX: LLC) - Real estate sector, growth potential
#10. APA Group (ASX: APA) - Utilities sector, stability and fundamentals.

These companies are selected based on their share price range, sector diversification, recent performance, fundamental strength, and growth potential.

Regarding your Amibroker AFL code for the Enhanced RSI Strategy
Your strategy looks well-structured. Here’s a brief overview of what it does:
  • Initial Setup: Sets the initial capital to $100,000 and limits the maximum number of open positions to 10.
  • Position Size: Each trade is set to $10,000.
  • Moving Average: Uses the Exponential Moving Average (EMA) over 27 periods.
  • RSI Strategy: Sets the RSI bounds between 40 and 60, and calculates the upper and lower bounds for buy and sell signals based on volatility.
  • Buy/Sell Signals: Generates buy signals when the current price is above the upper bound and sell signals when below the lower bound.
  • Stop Loss: Applies a 10% stop loss to manage risk.
This strategy aims to capture price movements within a defined range, using volatility-adjusted bounds to generate trading signals. If you have any specific questions or need further adjustments, feel free to ask!

Skate.
 
Skate said:
Summary
(AI) had no issue with the question or addressing the criteria listed above, and responded based on my input.
Click to expand...

Well, therein is the issue. You clearly have no undertsanding of how to be critical about the response.

Did you even notice that 80% of the results started with "A"? Do you even think this is reasonable? (Hint: Only 12% of ASX listed companies that are operating/holding companies begin with A)

You referenced "Adairs Limited (ADH)" and also specified "within ASX 200 Index" - ADH has never been in the S&P/ASX 200 index. It was a member of the S&P/ASX 300 from 2015-0921 through 2017-03-17 and rejoined the S&P/ASX 300 again on 2021-03-22.

You referenced "Alumina Ltd (AWC)". Did you even notice that was delisted a month ago?

You referenced "Arafura Resrouces Ltd (ARU)" - did you even know that it changed its named to Arafura Rare Earths Ltd almost 2 years ago?

You referenced "G.U.D. Holdings Ltd (GUD)" - this company changed its symbol to AOV 2 months ago, and changed its name to "Amotiv Limited"

There might be more errors, but 40+% error rate on your "simple question". Oh my. GTFO.

Trusting the output of LLMs is throwing darts but I bet you can only show a (single) result set that has superior returns and convince the luddites that this is the future.

Trusting AI output is the luddite future.

Data-based critical analysis of the output of AI to determine and exploit the flaws is the field of intellectuals.
 
Last edited:

The future of (AI)
@Richard Dale thank you for checking and posting the responses from (AI).

Your statement sums up (AI)
"Asking any AI engine about which stocks to choose is probably no better than a dartboard."

Skate.
 
Learning Opportunity
The feedback from @Richard Dale provides a valuable learning opportunity regarding the use of (AI) in trading. Richard’s comments emphasise the critical importance of data verification and thorough analysis when working with AI-generated outputs.

Clarifications:
1. The first question and response allowed the (AI) model to select companies based on a set of simple criteria.
2. The second question required the (AI) model to follow specific criteria provided in an Amibroker AFL (AmiBroker Formula Language).

Skate.
 
@Skate Professor do you really think that the constant criticism from one R Dale is at all constructive to a beginner who would more than likely get lost after the first sentence.
Though I have been trading for about 40 years, I find your posts based for a beginner very informative.
R Dale just seems intent on trying to destroy your positiveness.
 
Thanks for being the voice of reason (great example of the critical thinking this thread has lost).

In my honest opinion, I think Skate is stuck in the beginners cycle (jumping from one strategy/style/system to another - no long term follow through - no statistically significant backtests).
 
Skate said:
My Question
Could Artificial Intelligence (AI) be the answer for us as well to enhance our trading?
Click to expand...

in my opinion ... NO

but i live in contrarian world , but maybe i could use 'a reverse selection' system and THEN cull from there ( some stressed stocks are very likely to continue downhill )

now trend-traders might easily benefit , if AI detects increased broker interest in certain stocks ( say the AI monitors attendances of company presentations and correlates with opening new positions )

there are diverse opinions on 'broker recommendations ( especially if you aren't on their paid subscription list )

swing-traders might benefit as well by locating stocks liable to make quick reversals
 
i note the AI is using out-of-date data

AWC is now a take-over target ( in fact i believe actually taken-over )

BOQ given the recent results .. SAY WHAT ??

GUD .. hasn't that had a name/ticker code change ?

on the positive side , the flawed output might alert novices trying to stock-pick with AI that it is only a tool to begin researching/refining and not the complete answer
 

Embrace the Feedback
@farmerge, criticism, whether constructive or otherwise, can be tough to hear at times. However, in a forum like this, especially in this thread, we encounter a variety of views. When I post, some may find my musings helpful, others might ponder over the ideas presented, while some may not find any value at all. This diversity of opinions is natural and encouraged.

The “Dump it here” thread is designed for beginners who are on their journey to understand the markets. It’s a stepping stone, offering different perspectives and insights as they continue on their journey.

Skate.
 
You must log in or register to reply here.

Similar threads

R
My strategy/plan going forward
2
Replies
38
Views
2K
RbNEO
R
Very nice Rain here today
4 5 6
Replies
105
Views
5K
farmerge
  • Poll
Financial Independence (Home Ownership, Super, Long Retirement) not possible for wage earners: What do we need and how do we get it?
11 12 13
Replies
258
Views
21K
Smurf1976
November 2024 DDD
2 3
Replies
53
Views
6K
ducati916
Ozempic: Magic Pill or Devils Bargain?
Replies
10
Views
2K
sptrawler
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more...