Large language models (LLMs) are increasingly used as user simulators, but can they faithfully reproduce an individual's behavior and decision logic over time? We investigate this question in a longitudinal paper-trading study with 80 participants. Using rolling next-day predictions, we evaluate trading activity, buy-sell structure, asset selection, and portfolio outcomes, complemented by controlled input ablations and participant-rated explanation alignment. No evaluated LLM reliably outperforms a simple recent-activity persistence baseline for predicting trade occurrence, while buy-sell structure and asset selection remain difficult to reproduce. Similar activity-level performance can also yield substantially different portfolio trajectories. Ablations show that activity predictions depend primarily on recent trades, whereas asset choices are sensitive to multiple evidence sources. Participants rate explanations as more aligned with their recurring strategies than with their decision-specific rationales. These findings reveal a gap between coarse behavioral agreement and individual-level fidelity, motivating separate evaluation of behavioral accuracy, explanation alignment, and evidence dependence in personalized user simulation.
Figures & tables
Figure 1: Rolling investor simulation. Evidence available at market close on day t informs predictions for the next trading day t+1 ; newly observed records update subsequent inputs.
Trade Occurrence
Side Structure
Model
Acc.
Macro-F1
Trade Recall
Acc.
Macro-F1
Gemini-3-Flash
68.60
65.34
76.55
42.02
37.88
Gemini-3.5-Flash
75.06
68.94
61.89
36.48
34.05
GPT-4o-mini
76.11
69.32
58.63
33.88
35.80
GPT-5.6-Luna
78.05
71.23
59.28
34.53
34.53
GPT-5.6-Sol
79.18
66.24
34.85
23.13
28.46
Table 1: Trade-occurrence and end-to-end buy–sell structure prediction (%). Activity metrics use 1,239 scored participant–days; side metrics use all 307 observed trading days, counting no-trade predictions as incorrect. Bold indicates the highest point estimate per metric.
Figure 2: Activity and action-structure mismatch. (a) Predicted versus observed trading frequency. (b) Observed trading days partitioned into correct side predictions, incorrect side predictions, and missed trades. (c) Cumulative observed and predicted trading activity.
Model
Buy Hit@5
Buy NDCG@5
Sell Hit@5
Sell NDCG@5
Gemini-3-Flash
40.7
0.259
42.9
0.272
Gemini-3.5-Flash
39.0
0.260
42.9
0.265
GPT-4o-mini
37.7
0.247
27.8
0.164
GPT-5.6-Luna
35.2
0.223
42.9
0.257
GPT-5.6-Sol
22.9
0.150
32.3
0.219
DeepSeek-V4-Flash
25.8
0.164
35.4
0.227
Table 2: Candidate-filtered asset retrieval on all observed trading days, without conditioning on prediction correctness. Hit@5 is a percentage; NDCG@5 uses a 0–1 scale. Bold indicates the best result per column.
Figure 3: Illustrative comparisons of observed and simulated trading activity and portfolio trajectories. Similar activity-level fidelity can yield different portfolio outcomes.
Withheld evidence
Δ Acc. (pp)
Activity flip
Ticker flip
Recent trades
-23.49
36.5%
76.6%
Portfolio
-1.40
7.0%
58.6%
7-day behavior
-0.23
4.4%
56.8%
Market context
-1.16
4.0%
57.7%
Long-term profile
-0.23
6.7%
60.4%
Table 3: Evidence ablation with GPT-5.6-Sol. Accuracy changes are relative to the paired full-input condition. Activity flips indicate changes in trade-occurrence predictions; ticker flips indicate changes in either ordered buy or sell list.
Dimension
1
2
3
4
5
Mean
Strategy alignment
8
3
24
39
6
3.40
Decision-rationale alignment
18
22
28
12
0
2.43
Table 4: Participant-rated explanation alignment ( n=80 ; one explanation per participant). Ratings range from 1 to 5; higher indicates stronger perceived alignment.
Figure 4: Research escalation predicts subsequent trading, but placebo tests limit causal interpretation.
Appendix figures & tables3 assets
Supplementary material from the paper’s appendix.
Appendix
Quantity
Count
Prediction instances
1,239
Trading days
307
Non-trading days
932
Buying days
236
Selling days
198
Buy-and-sell days
127
Appendix
Table 5: Coverage of prediction and evaluation sets.
Model
Trade P
No-trade P
Trade R
No-trade R
Buy-only R
Sell-only R
Buy-and-sell R
Gemini-3-Flash
42.57
89.52
76.55
65.99
39.45
4.23
65.35
Gemini-3.5-Flash
49.74
86.35
61.89
79.40
21.10
4.23
67.72
GPT-4o-mini
51.58
85.73
58.63
81.87
30.28
7.04
51.97
GPT-5.6-Luna
55.32
86.26
59.28
84.23
23.85
5.63
59.84
GPT-5.6-Sol
64.85
81.38
34.85
93.78
11.01
8.45
41.73
DeepSeek-V4-Flash
54.94
82.21
41.69
88.73
11.01
7.04
50.39
Appendix
Table 6: Detailed User Simulator Behavioral Metric Evaluation(%). P and R denote precision and recall.
Evidence
Observed Information
Prediction setup
Cutoff: Aug. 11, 2026 (market close); Target: Aug. 12, 2026.
User profile
shown in Appendix B
Platform activity
333 interactions across 7 active days. Most viewed: RKLX (36 views), NEBX (18 views).
Recent trading
Traded on all 7 recent sessions; 31 buy and 42 sell executions. Both buying and selling occurred on all 7 sessions.