IB Personal Project (MYP) External Assessment Exemplar

Effective demonstration of critical thinking by evaluating “Holy Grail” trading indicators and refocusing on market fundamentals

Strategic use of decomposition to break the complex AI modelling goal into manageable sub-tasks, improving project manageability

Strong information literacy in sourcing, integrating, and evaluating code snippets from diverse references

Resourceful use of AI tools to clarify complex concepts, directly enhancing model sophistication

Organizational skills are described in data gathering, but the link between these skills and measurable improvements in model performance is not fully evidenced

Clearly defined, focused, measurable learning goal linking AI predictive modelling techniques with stock market trading

Engaging personal narrative that connects early coding fascination to current project motivation

Precisely stated product goal for a machine-learning-powered S&P 500 predictor

Proactive integration of research at every project stage to anticipate and mitigate challenges such as data leakage and overfitting

Comprehensive long-term Gantt chart with clear sequencing, durations, and resource planning

Detailed short-term Trello board plan breaking tasks into actionable cards with deadlines

Overly dense consolidated specifications table lacking concise, measurable success benchmarks

Repetitive restatement of the eight-step process in multiple sections, which reduces flow and focus

Incomplete entries in the first specifications table (missing research justification and criteria)

Success criteria tables (especially for cost & materials) with missing threshold values and unclear, non-quantifiable indicators

Overall tables could be reorganized or broken into smaller sub-tables for clarity and readability

Nuanced discussion of backtesting versus live-testing limitations, reflecting deep methodological awareness

Mature concluding reflection that acknowledges both achievements and challenges, demonstrating self-evaluation

Clear articulation of personal growth in quantitative analysis, programming, and financial understanding

Rigorously evidence-based evaluation using quantitative backtesting and live-testing metrics

Transparent presentation of precision_score calculations to support model accuracy claims

Balanced perspective on model stability by highlighting limited real-world testing timeframe

Second evaluation table contains missing results and blank cells, undermining comprehensive assessment

Use of descriptive labels for “Level of Success” rather than numeric rubric levels reduces clarity of criterion alignment