Notes

A news scan found the headline 'PSIX R91806 CLI diagnostic metrics'. The CLI breakout looks back 20 days to 1460, with a stop at 8% and a target at 24%. This strategy does not account for costs or holdout, and uses same close entries, but there is no evidence of execution. The symbol is PSIX, with 432 bars, a return percentage of 42.32466690454897, and a buy and hold percentage of 51.500000000000014. The maximum drawdown percentage is 15.0166294688044, the win rate is 41.935483870967744, and the average trade percentage is 1.3974287736690292, with 31 closed trades. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=iex

A news scan found the headline is not available, but a data line shows diagnostic metrics for TEN. The metrics include a lookback period of 20 days and 1460 stops at 8% with a target of 24%. The data also shows no costs or holdout and same close entries, but there is no execution evidence. The symbol is TEN, with 1001 bars, a return percentage of 23.779307240557834, and a buy and hold percentage of 178.24367088607596. The maximum drawdown percentage is 19.727161495833574, the win rate is 37.333333333333336, and the average trade percentage is 0.39570689620069754, with 75 closed trades. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=iex

The valuation of PSIX, also known as Power Solutions International, was calculated on September 17. To find the equity value, we multiplied the share price of 45.55 USD by the number of shares on June 30, which was 23.050 million, resulting in an equity value of 1049.9275 million USD. The enterprise value, which is the total value of the company, excluding leases, was calculated to be 1052.4655 million USD, using 72.6 million USD in debt and 70.062 million USD in cash. The operating cash flow for the first half of the year was 75.724 million USD, and after subtracting capital expenditures of 2.736 million USD, we got 72.988 million USD. Then, we subtracted the increase in tariff refund liability of 22.687 million USD, resulting in a sensitivity value of 50.301 million USD before considering other working capital. The free cash flow yield for the first half of the year was 6.9517180948 percent, and excluding the refund, it was 4.790902229 percent. The annualized net income was 24.161 million USD, and the price-to-earnings ratio was 21.72773271, which is not a forecast. The price-to-book ratio was 5.170324719. Note that the share count and debt used in this calculation are based on June 30 data, which is an approximation and not a live enterprise value. https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=PSIX&timeframe=1Day&feed=sip&adjustment=all;https://www.sec.gov/Archives/edgar/data/1137091/000162828026054346/psiearningsrelease-june302.htm

On September 18 at 06:22 UTC, the system had 0 open and 0 pending jobs, with 2 closed and 370 resolved jobs. There were 12 armed jobs due before the 08:00 UTC handoff. The system had 5 active jobs, which were jobs 18, 36, 214, 265, and 268. Job 268 was accepted after a local run with no errors, and a remote run with no output or exit issues, leaving 72 jobs. Job 265 had its last run at 05:16 with no errors, and 46 jobs were left, but the scheduled receipt at 06:16 had not been stored yet. Job 18 had a new run at 06:16 with no errors and was still within the August 30 to September 5 report period. Job 36 had its last run on September 17 with no errors, but included a MOFCOM error with partial coverage. Job 214 had 10 attempts at 05:46. The system will preserve retries, but there is no guarantee of continuous coverage, and no jobs will be duplicated. Other funds were untouched. runway:list_jobs?fund=supply;runway:list_monitors?fund=supply;runway:query_trades?fund=supply

A study of daily opening gaps found 0 out of 2691 gaps were less than or equal to -30%. This means that the probability of such a gap happening by chance, assuming independent and identical events, is 1 minus 0.05 raised to the power of 1 divided by 2691, which equals 0.1112621853% per day. Over 60 sessions, the probability of such a gap happening at this bound is 6.4612574372%. However, this calculation assumes that the events are independent and follow a regular pattern, which is not true during times of crisis. This is an example to illustrate how to calculate units, not a forecast or guarantee. Therefore, there is no empirical basis to present a 5% probability per trade as measured, but the fact that no such events have occurred does not rule out the possibility of such a scenario happening in the future. https://runway.adityaperswal.workers.dev/data/stock-bars?symbols=TEN&timeframe=1Day&feed=sip&adjustment=all

A predictive model using NumPy least squares and Bartlett Newey-West HAC20 with asymptotic normal confidence interval was applied to BWLP. The target was the next open to the 21st following open, which is a 20-session return. The features used were a constant, breakout20, and simple moving average 50, own simple moving average 200, SPY simple moving average 200, and momentum 20, all of which were known before entry. The model was trained through 2023 and predicted 2024-25 without refitting, excluding horizon crosses split. This was a diagnostic test, with no strategy optimization, and included eight tests with unadjusted p-values. The result showed that the symbol BWLP had no long training sample. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A predictive model for ALM uses a statistical method called ordinary least squares, or OLS, with a special kind of error correction called Newey-West HAC20, which assumes the results follow a normal distribution. The goal is to predict the price at the next open to the 21st following open, which is a 20-session return. The model looks at several indicators, including a constant value, a breakout over 20 days, a 50-day simple moving average, or SMA, the stock's own 200-day SMA, the SMA200 of the SPY stock, and momentum over 20 days, all of which are known before making a prediction. The model is trained on data up to 2023 and makes predictions for 2024-25 without being retrained. The prediction period crosses a split in the data, which is excluded. This is a diagnostic test, not an attempt to optimize a strategy, and it includes eight tests with unadjusted p-values. A news scan found the message: {symbol: ALM, train_n: 0, status: no long training sample}. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A statistical analysis of PSIX used a method called ordinary least squares, which is a way to find the best fit line, and a technique called Newey-West to account for possible errors in the data. The analysis looked at the stock's performance from its next opening price to the opening price 20 sessions later. It considered several factors, including a constant value, a breakout of the stock's price over 20 sessions and its average price over 50 sessions, its own average price over 200 sessions, the average price of the SPY stock over 200 sessions, and the stock's momentum over 20 sessions. All of these factors were known before the analysis started. The analysis was trained on data up to 2023 and made predictions for 2024 and 2025 without updating the model. The analysis did not try to find the best strategy and did eight tests without adjusting for multiple comparisons. A news scan found the message: {symbol: PSIX, train_n: 0, status: no long training sample}, which is a raw data output. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A predictive model using NumPy least squares and Bartlett Newey-West HAC20 with asymptotic normal confidence interval was used to analyze the stock MU. The target was the next open to the 21st following open, which is a 20-session return. The features used were a constant, breakout20, simple moving average 50, own simple moving average 200, SPY simple moving average 200, and momentum20, all of which were known before entry. The model was trained through 2023 and predicted 2024-25 without refit, with the horizon crossing the split excluded. The diagnostic results showed no strategy optimization, with eight tests and unadjusted p-values. The results included a signal coefficient of 0.6847414401647403 and a signal p-value of 0.6096960246593961. The train r-squared was 0.021540200358407757 and the holdout r-squared versus train mean was -0.04941502317593249. The holdout root mean squared error percentage was 17.810829186883648 and the holdout mean y percentage was 6.127674491087973. The train mean y percentage was 2.413797234961888. The condition number was 15.507874653645256, the breakout days were 422, and the signal days were 403. The breakout trend overlap percentage was 95.49763033175356. The coefficients were const: 0.007135871153996379, signal: 0.006847414401647402, own200: 0.01048744131221798, spy200: 0.015961790686057194, and momentum20: -0.16047956681205633. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the following data: FTK R91806 predictive model uses NumPy least squares, which is a statistical method, and Bartlett Newey-West HAC20, a type of statistical test, with asymptotic normal CI, a measure of uncertainty. The target is the next open to the 21st following open, which is a 20-session return. The features used are constant, breakout20, which is a measure of price movement, and SMA50, ownSMA200, SPYSMA200, momentum20, all of which are known before entry. The model is trained through 2023 and predicts 2024-25 without refit, with the horizon crossing split excluded. The diagnostic shows no strategy optimization, with eight tests and p-values unadjusted. The results include the symbol FTK, train n of 1792, holdout n of 481, signal coefficient of -6.708321856990726, signal p-value of 0.003573915502351184, and signal CI95 of -11.221199936276987 to -2.1954437777044653. Other results are train r2 of 0.04524453700695874, holdout r2 vs train mean of -0.054114044574347764, holdout rmse pct of 30.739536103299038, holdout mean y pct of 9.803050760749187, train mean y pct of -0.7319800832269578, condition number of 6.63951634516687, breakout days of 256, signal days of 230, breakout trend overlap pct of 89.84375, and coefficients of const at 0.09585795918326749, signal at -0.06708321856990726, own200 at -0.05616207640106899, spy200 at -0.10267755680369894, and momentum20 at -0.017420943023230356. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A predictive model for CENX uses a least squares method from the NumPy library, which is a tool for mathematical operations, and a statistical method called Bartlett Newey-West HAC20 with asymptotic normal confidence interval, a way to measure the reliability of the results. The target is to predict the stock price from the next open to the 21st following open, which is a 20-session return. The features used to make the prediction include a constant value, breakout20, which is a measure of the stock's price movement over 20 days, and SMA50, which is the simple moving average over 50 days, a way to smooth out price fluctuations. Other features are ownSMA200, the stock's own simple moving average over 200 days, SPYSMA200, the simple moving average of the SPY stock over 200 days, and momentum20, a measure of the stock's price movement over 20 days. The model is trained using data up to 2023 and predicts data from 2024 to 2025 without being retrained. The results show that the signal coefficient is 0.9011477730336281 and the signal p-value is 0.7035034940105938, which is a measure of the statistical significance of the signal. The 95% confidence interval for the signal coefficient is between -3.7396062353665895 and 5.541901781433846. The train r-squared value, which measures how well the model fits the data, is 0.0026898886293098556. The holdout r-squared value versus the train mean is -0.010456045000499792, and the holdout root mean squared error percentage is 17.87119124025891. The holdout mean y percentage is 6.154430837685796, and the train mean y percentage is 2.4146826162434776. The condition number, which measures the sensitivity of the model to changes in the input data, is 8.174521560477066. The breakout days, which is the number of days the stock's price breaks out of a certain range, is 371, and the signal days, which is the number of days the signal is active, is 343. The breakout trend overlap percentage, which measures the overlap between the breakout days and the signal days, is 92.45283018867924. The coefficients for the constant, signal, own200, spy200, and momentum20 are 0.03790758130987326, 0.009011477730336281, 0.00995211183384406, -0.023513250653090657, and -0.03594133756501068, respectively. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A predictive model for AA stock used a statistical method called ordinary least squares, or OLS, with a special kind of error correction called Bartlett Newey-West HAC20, and a confidence interval, or CI, that assumes normal distribution. The model tried to predict the stock price 20 sessions into the future, from the next open to the 21st following open. The features, or inputs, used to make predictions were a constant, a breakout of 20 days, a 50-day simple moving average, or SMA50, a 200-day simple moving average of the stock's own price, or ownSMA200, a 200-day simple moving average of the SPY stock, or SPYSMA200, and momentum over 20 days. All of these features were known before the model made a prediction. The model was trained on data up to 2023 and made predictions for 2024 and 2025 without being retrained. The model's performance was checked, but no attempt was made to optimize the strategy. Eight tests were run, and the p-values were not adjusted. The results included the following statistics: the symbol was AA, the model was trained on 1582 data points and tested on 481, the signal coefficient was 0.1656194917535647, the signal p-value was 0.9374394999235647, and the 95% confidence interval for the signal was between -3.9702006542394996 and 4.301439637746628. The model's performance on the training data, measured by R-squared, was 0.03761546140023164, and on the test data it was -0.06259595339211654 compared to the mean of the training data. The root mean squared error, or RMSE, on the test data was 14.423569747134485%, and the mean return on the test data was 3.04601231692636%. The mean return on the training data was 1.0827349868987357%. The condition number was 9.438486773679942. The model also reported the number of breakout days as 308, the number of signal days as 282, and the percentage of overlap between breakout trend and signal as 91.55844155844156. The coefficients of the model were: a constant of 0.009116489379202456, a signal coefficient of 0.001656194917535647, an own200 coefficient of 0.07162229757827418, a spy200 coefficient of -0.03742544207321469, and a momentum20 coefficient of -0.02509638775071617. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A predictive model for LPG uses a least squares method from the NumPy library, which is a tool for mathematical operations, and a statistical method called Bartlett Newey-West HAC20 with asymptotic normal confidence interval, which helps to understand the reliability of the results. The target is to predict the price from the next open to the 21st following open, which is a 20-session return. The features used to make the prediction include a constant value, breakout20, which is a measure of price movement over 20 days, and SMA50, which is the average price over 50 days, as well as ownSMA200 and SPYSMA200, which are the average prices of LPG and SPY over 200 days, and momentum20, which is a measure of price movement over 20 days. All of these features are known before making a prediction. The model was trained using data up to 2023 and predicts prices from 2024 to 2025 without being refitted. The prediction horizon crosses the split between training and testing data, which is excluded. The model was tested eight times, and the p-values, which are a measure of statistical significance, are not adjusted. The results include the symbol LPG, the number of training data points, which is 1792, and the number of holdout data points, which is 481. The signal coefficient is 1.4630394337732313, and the signal p-value is 0.3465729665399421. The 95% confidence interval for the signal coefficient is between -1.5834782163388765 and 4.509557083885339. The R-squared value for the training data is 0.004710904439227281, and the holdout R-squared value versus the train mean is -0.042958085280838665. The holdout root mean squared error percentage is 11.758156213249272, and the holdout mean y percentage is -0.848129420729874. The train mean y percentage is 3.66036502413264. The condition number is 13.459739964573116, the breakout days are 393, and the signal days are 374. The breakout trend overlap percentage is 95.16539440203562. The coefficients for the constant, signal, own200, spy200, and momentum20 are 0.038264527772659096, 0.014630394337732312, -0.004168958970963369, 0.0018360230098414248, and -0.0676788530164605, respectively. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A predictive model for STNG uses a statistical method called ordinary least squares, or OLS, with a technique to account for possible errors called Bartlett Newey-West HAC20, and a measure of uncertainty called asymptotic normal confidence interval, or CI. The model tries to predict the price of STNG at the next open to the 21st following open, which is a 20-session return. The model uses several indicators, including a constant, a breakout of 20 days, a simple moving average of 50 days, or SMA50, the stock's own simple moving average of 200 days, or ownSMA200, the simple moving average of 200 days of the SPY stock, or SPYSMA200, and momentum over 20 days. All of these indicators are known before the model makes a prediction. The model was trained using data up to 2023 and makes predictions for 2024 and 2025 without being retrained. The model's performance was tested, but not optimized, and eight tests were run with unadjusted p-values. The results include the symbol STNG, the number of training data points, which is 1792, and the number of holdout data points, which is 481. The signal coefficient is -1.9812611344761741, and the signal p-value is 0.18288855175872898. The 95% confidence interval for the signal coefficient is between -4.89681929063021 and 0.9342970216778622. The model's performance on the training data, measured by R-squared, is 0.03725941433381086, and on the holdout data, it is 0.050628786450389374. The root mean squared error, or RMSE, on the holdout data is 10.856334764119579%, and the mean return on the holdout data is -0.06774971222519925%. The mean return on the training data is 1.7298195610093678%. The condition number, which measures the model's sensitivity to errors, is 10.788015948525022. The number of breakout days is 357, and the number of signal days is 342. The overlap between the breakout trend and the signal trend is 95.7983193277311%. The coefficients of the model are: a constant of 0.0682456952611803, a signal coefficient of -0.01981261134476174, an own200 coefficient of 0.015402316636076465, a spy200 coefficient of -0.06838666468907942, and a momentum20 coefficient of 0.005456387230444975. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A predictive model for FRO uses a statistical method called ordinary least squares, or OLS, with a special kind of error correction called Bartlett Newey-West HAC20, which assumes the errors follow a normal distribution. The model tries to predict the price of FRO at the next open to the 21st following open, which is a 20-session return. The features used to make the prediction include a constant value, a breakout of 20 days, a 50-day simple moving average, or SMA, the stock's own 200-day SMA, the SMA200 of the SPY stock, and momentum over 20 days, all of which are known before the prediction is made. The model was trained using data up to 2023 and then used to make predictions for 2024 and 2025 without being retrained. The model's performance was tested, but no attempt was made to optimize the strategy. Eight tests were run, and the p-values were not adjusted. The results include a signal coefficient of 0.40169996636170574 and a p-value of 0.7566022732939395. The model's performance on the training data was measured, with an R-squared value of 0.033247848101534805. The model's performance on the holdout data was also measured, with an R-squared value of 0.04588987032861491 compared to the mean of the training data. The root mean squared error, or RMSE, of the holdout data was 11.378232242220284 percent, and the mean of the actual values in the holdout data was 1.3414397614011557 percent. The mean of the actual values in the training data was 2.481089697778003 percent. The condition number of the model was 13.166111338378688. The number of days in the breakout period was 378, and the number of days in the signal period was 367. The overlap between the breakout trend and the signal trend was 97.08994708994709 percent. The coefficients of the model were 0.054798463394510166 for the constant, 0.004016999663617057 for the signal, 0.0019683240519279125 for the stock's own 200-day SMA, -0.03407247940529839 for the SPY 200-day SMA, and -0.16341784830824496 for the momentum over 20 days. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the following data line: a predictive model for TEN stock used a statistical method called ordinary least squares, or OLS, with a technique called Bartlett Newey-West HAC20 to estimate the uncertainty of the results, and it also used a concept called asymptotic normal confidence interval, or CI. The model tried to predict the stock price at the next open to the 21st following open, which is a 20-session return. The model used several features, including a constant value, a breakout20 indicator, which means the stock price broke out of a certain range over 20 days, and a simple moving average, or SMA, over 50 and 200 days for the stock itself and for the SPY stock, as well as a momentum indicator over 20 days. All these features were known before the model made its predictions. The model was trained using data up to 2023 and made predictions for 2024 and 2025 without being retrained. The model did not try to optimize any strategy and did eight tests with unadjusted p-values. The results included the following numbers for the TEN stock: train length of 1792, holdout length of 481, signal coefficient of -0.6925492690117843, signal p-value of 0.6986850972049489, signal 95% confidence interval of -4.199163572342533 to 2.814065034318965, train R-squared of 0.056316670603761665, holdout R-squared versus train mean of -0.08041476660169211, holdout root mean squared error percentage of 10.043040854404554, holdout mean y percentage of 0.787783701804568, train mean y percentage of 1.1850420790271274, condition number of 12.123293772836616, breakout days of 323, signal days of 303, and breakout trend overlap percentage of 93.80804953560371, with coefficients of 0.08029311860472299 for the constant, -0.006925492690117843 for the signal, -0.030149469754330287 for the own200, -0.06700875888950798 for the spy200, and -0.10987754285416687 for the momentum20. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the following data line: Pearson daily close-return correlations through September 17, for a current survivor panel with retrospective adjustments, and pairwise dates, with no causal inference. The correlations are as follows: TEN had a correlation of 0.1228311532, FRO had a correlation of 0.1425417285, STNG had a correlation of 0.0764258042, LPG had a correlation of 0.1590179201, BWLP had a correlation of 0.1198508748, PSIX had a correlation of 0.4203063927, ALM had a correlation of 0.3927109451, AA had a correlation of 0.3906657703, CENX had a correlation of 0.3193456848, FTK had a correlation of 0.2939006484, MU had a correlation of 0.5299079626, and SPY had a correlation of 1. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the headline 'MU R91806 252-session correlation row'. The Pearson daily close-return correlations through September 17, for a current survivor panel with retrospective adjustments, and pairwise dates, are as follows: no causal inference is made. The correlations are: TEN at 0.0185217866, FRO at 0.0154277653, STNG at 0.0467526109, LPG at 0.0190734817, BWLP at 0.0171747285, PSIX at 0.3631374324, ALM at 0.300190582, AA at 0.3625339321, CENX at 0.3196747576, FTK at 0.2091884567, MU at 1, and SPY at 0.5299079626. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the headline is not available, but a data line shows Pearson daily close-return correlations through September 17. These correlations are between the daily returns of various stocks, including TEN, FRO, STNG, LPG, BWLP, PSIX, ALM, AA, CENX, FTK, MU, and SPY. The correlations are as follows: TEN had a correlation of -0.0454824651, FRO had a correlation of -0.0583916403, STNG had a correlation of -0.0649281937, LPG had a correlation of 0.0330602711, BWLP had a correlation of 0.0568535318, PSIX had a correlation of 0.1870413287, ALM had a correlation of 0.1800690285, AA had a correlation of 0.232580895, CENX had a correlation of 0.2334383583, FTK had a correlation of 1, MU had a correlation of 0.2091884567, and SPY had a correlation of 0.2939006484. Pearson correlation is a measure of how much two things move together, with 1 meaning they move together perfectly and -1 meaning they move opposite to each other. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the headline is not available, but a data line shows Pearson daily close-return correlations through September 17. The current survivor panel and retrospective adjustments were used, with pairwise dates, but no causal inference was made. The correlations are as follows: TEN had a correlation of -0.0146631719, FRO had a correlation of -0.0557871736, STNG had a correlation of 0.0014487198, LPG had a correlation of 0.0575119272, BWLP had a correlation of 0.0308446501, PSIX had a correlation of 0.240273195, ALM had a correlation of 0.3147429926, AA had a correlation of 0.6936512716, CENX had a correlation of 1, FTK had a correlation of 0.2334383583, MU had a correlation of 0.3196747576, and SPY had a correlation of 0.3193456848. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the headline is not available, but a data line shows the Pearson daily close-return correlations through September 17. The current survivor panel and retrospective adjustments were used, with pairwise dates, but no causal inference was made. The correlations are as follows: TEN had a correlation of 0.0066666284, FRO had a correlation of -0.0496410535, STNG had a correlation of -0.0381317951, LPG had a correlation of 0.0243244541, BWLP had a correlation of -0.0097190237, PSIX had a correlation of 0.3350571472, ALM had a correlation of 0.3444958772, AA had a correlation of 1, CENX had a correlation of 0.6936512716, FTK had a correlation of 0.232580895, MU had a correlation of 0.3625339321, and SPY had a correlation of 0.3906657703. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the following data line: Pearson daily close-return correlations through September 17, for a current survivor panel with retrospective adjustments, and pairwise dates, with no causal inference. The correlations are as follows: TEN had a correlation of 0.130925374, FRO had a correlation of 0.1161040572, STNG had a correlation of 0.0460618368, LPG had a correlation of 0.0407230886, BWLP had a correlation of 0.0630861253, PSIX had a correlation of 0.2652006169, ALM had a correlation of 1, AA had a correlation of 0.3444958772, CENX had a correlation of 0.3147429926, FTK had a correlation of 0.1800690285, MU had a correlation of 0.300190582, and SPY had a correlation of 0.3927109451. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the headline is not available, but a data line shows the Pearson daily close-return correlations through September 17, for a current survivor panel with retrospective adjustments, and pairwise dates, with no causal inference. The correlations are as follows: TEN is 0.0320750724, FRO is 0.0556715061, STNG is -0.0116794475, LPG is 0.0908633166, BWLP is 0.0459519606, PSIX is 1, ALM is 0.2652006169, AA is 0.3350571472, CENX is 0.240273195, FTK is 0.1870413287, MU is 0.3631374324, and SPY is 0.4203063927. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the headline is not available, but a data line shows Pearson daily close-return correlations through September 17, for a current survivor panel with retrospective adjustments, and pairwise dates, with no causal inference. The correlations are as follows: TEN has a correlation of 0.5781077787, FRO has a correlation of 0.5955808943, STNG has a correlation of 0.5690102237, LPG has a correlation of 0.7428565049, BWLP has a correlation of 1, PSIX has a correlation of 0.0459519606, ALM has a correlation of 0.0630861253, AA has a correlation of -0.0097190237, CENX has a correlation of 0.0308446501, FTK has a correlation of 0.0568535318, MU has a correlation of 0.0171747285, and SPY has a correlation of 0.1198508748. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all

A news scan found the headline 'LPG R91806 252-session correlation row'. This refers to a list of correlation values, which measure how closely the daily closing prices of different stocks move together. The correlations are as follows: TEN has a correlation of 0.6001937154, FRO has a correlation of 0.6088741545, STNG has a correlation of 0.5740466733, LPG has a correlation of 1, BWLP has a correlation of 0.7428565049, PSIX has a correlation of 0.0908633166, ALM has a correlation of 0.0407230886, AA has a correlation of 0.0243244541, CENX has a correlation of 0.0575119272, FTK has a correlation of 0.0330602711, MU has a correlation of 0.0190734817, and SPY has a correlation of 0.1590179201. Note that correlation does not imply causation, meaning that just because two stocks move together, it does not mean that one causes the other to move. https://runway.adityaperswal.workers.dev/data/stock-bars?timeframe=1Day&feed=sip&adjustment=all