**NOTICE: Currently experiencing issues with our datasources. Report generation will not work properly - please try later.**
Search Results (Select)
NameSymbolExchangeType
*Note: Most accurate predictions will be based on Canadian markets only, as the model is trained on Canadian
economic data.
Your Portfolio
Use the search bar to add your first security!
Balanced
Low Risk Balanced High Returns
Lambda Value - λ
The selected λ (lambda) controls how the optimization balances risk and return. The model evaluates each portfolio using the following score:
Here, Expected Return is predicted by the neural network, and Spread represents the width of the 90% confidence interval (proxy for risk).
A higher λ places more weight on expected return, favoring portfolios with greater upside potential even if they are more risky. A lower λ emphasizes
stability by prioritizing portfolios with tighter confidence intervals, even if expected returns are lower.
The optimization seeks to maximize this score, selecting allocations amoungst securities that best align with your chosen risk/return preferences.
A good portfolio will find a balanced between these two characteristics.
Current queue: Loading...
Approx. wait time per report ~ 2mins
Building Your Dashboard
Approximate wait time: ~2mins
Waiting for job to start...
My Dashboard - Optimization Results
*Disclaimer: This document is for research purposes only and does not constitute financial advice. Any
investment decisions based on its contents are made at the risk of the individual. The author disclaims any
responsibility for gains, losses, or other consequences.
Heads up! Chart visuals are available on desktop only. Use your PC for the full chart view.
00.00%
Yield Prediction
*Quarterly estimate
PoorGood
00.00%
90% Confidence Interval
*Quarterly estimate
GoodPoor
Recommended Portfolio
InvestmentWeight
Model Diagnostics
Statistical Summary
Pro-tipAdd more low-risk assets or decrease λ until reliability is ≥25/30.
*Disclaimer: This document is for research purposes only and does not constitute financial advice. Any
investment decisions based on its contents are made at the risk of the individual. The author disclaims any
responsibility for gains, losses, or other consequences.
Portfolio
InvestmentWeight (%)
*Adjust the ‘Weight’ and ‘Budget’ fields above to review your current portfolio or explore alternative allocations. Click ‘Regenerate’ to update the analysis on this page.
Expected Yield
0.00%
$0.00
*Last projected percentage from chart
90% Confidence Interval
0.00% ~ 0.00%
($0.00 ~ $0.00)
*Last projected percentage range from chart
Model Diagnostics
Statistical Summary
Pro-tipAdd more low-risk assets or decrease λ until reliability is ≥25/30.
This tab generates an optimal allocation from the securities you selected.
Technical details about the modelling process are available on the Behind the Model page.
Yield Optimization
This chart shows how the model converges toward a solution. The objective is to
maximize predicted yield while minimizing uncertainty (risk).
Yield Prediction & Confidence Interval
These boxes display expected performance and the 90% confidence interval along with relative ratings. This help show how the model expects the portfolio to perform, and how certain it is about that estimate.
Important: The model predicts quarterly performance (3-month horizon). All results refer to this period.
Recommended Portfolio
The table provides the allocation weights that produce the optimal solution.
All reported predictions correspond specifically to this recommended portfolio.
Prediction vs Real Performance
This chart evaluates how well the model predicts actual outcomes.
Orange line — Actual Performance: Realized yield over the previous 3 months.
Prediction line — Forecast: Uses only data prior to that performance window.
Example: If a point is plotted at 2025-09-25
Actual performance: 2025-06-25 → 2025-09-25
Prediction data: before 2025-06-25
This prevents information leakage and ensures a fair comparison.
The light-blue band around the prediction line represents model uncertainty.
Model Diagnostics
Provides a statistical summary of prediction accuracy.
Only overlapping periods (where both predicted and real values exist) are evaluated.
Manage Tab
This tab analyzes an existing portfolio.
Edit allocation weights in the Portfolio table (must sum to 100%)
Change the starting budget
Budget Note: To interpret projected returns as realistic outcomes,
the portfolio must be assumed to have been purchased at the beginning of the study period.
Dividend Assumption: Dividends are treated as cash and not reinvested.
Current Portfolio vs Expectation
Similar to the prediction chart in the Optimal tab, but shows
market value instead of percentage yield,
giving a clearer picture of wealth growth over time.
A diagnostics panel is included to evaluate prediction accuracy
against real portfolio value.
Updating the Portfolio
Modify weights or budget → click Regenerate
Add/remove securities → click Change Portfolio and update selections
Pro Tips
Since some portfolios are inherently more uncertain, add more low-risk securities to your portfolio to help diversify
and reduce overall risk. This helps stabilize returns and lowers volatility by offsetting the impact of higher-risk assets.
A good way to measure this is by ensuring the Model Reliability Score is greater than 25/30.
